00:00:00.000 - 00:00:22.560

Okay, good morning everybody. Could I ask the PCAST members to take their seats? And we will call this PCAST public plenary meeting to order. It is a pleasure once again to welcome the members of PCAST, the members of the OSTP

00:00:22.560 - 00:00:52.680

staff who are with us, and the members of the wider science and technology community in the public who have joined us both in person and on the live stream on the web. As usual, we have a full schedule this morning with forensics, near-earth objects, cryptocurrencies, and that is reflective of the full schedule that PCAST has had throughout its tenure in

00:00:52.680 - 00:01:19.800

the Obama administration, and that in turn is a reflection of the extraordinary degree of engagement and interest of President Obama in matters of science, technology, and innovation. I can tell you that nothing is slowing down as we go into the last 36 weeks, and maybe at the beginning of next week, now 35 weeks of this administration.

00:01:19.800 - 00:01:42.760

The President is determined to advance as far as possible as many of his signature initiatives as possible, and that includes the many that are in the science, technology, and innovation domain, and certainly PCAST has a number of studies underway that we want to bring across the finish line before the end of the term.

00:01:42.760 - 00:02:00.280

With that said, let me turn, without any further ado, to my PCAST co-chair, Dr. Eric Lander, who will add his welcome, but then we'll turn promptly to the topic of forensic science. Eric. John, thank you very much.

00:02:00.280 - 00:02:15.680

Let me again welcome all the members of PCAST, everyone here in the room, and everyone who is and will be watching on the web. There is a lot that is still going on at PCAST. Under today's meeting, we still have four more official meetings set for this group,

00:02:15.680 - 00:02:34.120

and the work is not letting up in any way. I'm really grateful to the members of PCAST for their active engagement in at least three ongoing studies and a few other projects, and so it's just great to have everyone's full participation.

00:02:34.120 - 00:02:53.320

I wanted to take a moment and update everyone about the project that PCAST has been engaged in on forensic science. It's a project we've mentioned before, and we're now nearing completion of the project, where the process of trying to write up a report on that.

00:02:53.320 - 00:03:13.280

I wanted to at least describe a little bit of motivation for it, the work underlying it, and directionally where we're going. We still haven't reached any conclusions yet, but I think it's pretty clear where we're going in terms of the questions that are being asked.

00:03:13.280 - 00:03:39.600

Forensic science in some sense is a very old discipline. Folks recognized more than a century ago that different types of scientific and technological evidence could be useful in the criminal court systems, fingerprints being an old example, but many different types of patterns that might be encountered in the crime scene might

00:03:39.600 - 00:04:07.080

be used to connect the evidence at a crime scene to a particular individual, a perpetrator or in some cases a victim. There were many different technologies that were proposed early in the 20th century. There were early forms of lie detectors that were offered up, and these lie detectors

00:04:07.080 - 00:04:28.880

really lacked much scientific foundation, and the courts in an important decision, federal pellet decision, began to erect some standards as to what it meant to have scientifically acceptable testimony in the courts. There's been a parallel development of methodologies and considerations of how to evaluate what

00:04:28.880 - 00:04:50.120

those methodologies are appropriate for use in courts. A real watershed moment occurred in the late 1980s, early 1990s with the introduction of DNA fingerprinting technology. Based on the scientific recognition that the human genome was variable between people and

00:04:50.120 - 00:05:12.600

that some regions of the human genome were quite variable, had many different alternative, we say alleles, different forms of the DNA, it was possible to find in medical research labs that if you took DNA from different people, you could observe very different patterns at particular places in the human genome.

00:05:12.600 - 00:05:35.960

In the late 1980s, this began to be introduced in courts as a forensic matching technology. You find some blood at the scene of the crime, and you match it to a perpetrator or a victim or you might find blood on a potential perpetrator and match it to a victim or blood at the scene of a crime and match it to a perpetrator, began to be introduced.

00:05:35.960 - 00:06:05.680

Very interestingly, despite being the most powerful of the forensic technologies ever developed, in its maiden voyages, it encountered some rough seas, and in some notable cases DNA evidence was excluded from the courts as being inappropriately practiced. This led to a tremendous effort to set standards for the use of DNA evidence, and within a

00:06:05.680 - 00:06:25.520

couple of years, through work of folks notably at the FBI's own laboratory, to DNA becoming a gold standard. It taught us something very important. It taught us that even the most absolutely superb basic science still has to be practiced

00:06:25.520 - 00:06:50.040

reliably to be useful in a forensic setting. This combination of scientific thinking and then real forensic work to set standards and assess reliability together produced an amazing gold standard that is used in the courts today. The DNA analysis, at least of single samples, unmixed samples, is really remarkably reliable.

00:06:50.040 - 00:07:09.080

But the minute we had a reliable tool, you could start using it to go back to other cases and ask, were there any mistakes made? And this became the next real revelation in forensic science that began to gather a little bit of steam in the 90s, but really it's been in the last decade or so that it's become

00:07:09.080 - 00:07:33.840

clear, which is DNA evidence has made it unambiguous that there are wrongful convictions. In fact, people have been sentenced to death wrongfully, and we know this now because we can in many cases see the evidence of a blood stain that was said to match someone that when you look at DNA, it clearly doesn't match the someone.

00:07:33.840 - 00:07:54.200

In fact, it often matches somebody else pointing to the true perpetrator. So we actually have two wrongs here, a wrong of an individual who's been wrongfully jailed and a wrong to society of leaving a true perpetrator on the street. Both are serious issues with regard to justice.

00:07:54.200 - 00:08:07.760

The ability to know we make mistakes is central to science. Science is about the ability to detect mistakes. Nothing is perfect. What sets science apart from some of their activities is we know it, we embrace it, we

00:08:07.760 - 00:08:24.600

measure it, and we can therefore get better because of it. So it has provoked some real thinking about how did the criminal justice system get things wrong? Well, in more than half of the cases, roughly half of the cases, the evidence included scientific

00:08:24.600 - 00:08:45.840

forensic science evidence that contributed to a conviction, and because it made a claim that we now know was wrong, we know it made a false claim. How did that happen? So that's pointed to problems with technologies like bite marks or even fingerprints or hair

00:08:45.840 - 00:09:08.840

analysis or other types of things. And what it provides is a great opportunity to increase the quality of forensic science. The National Academy in 2009 issued a report from an NRC committee that described what can only be said to be deeply serious problems with forensic science.

00:09:08.840 - 00:09:34.120

That report concluded that there were issues with the underlying scientific validity of many of these areas, and it made a whole series of recommendations. That report really has been tremendously important in focusing people's attentions on the problem. The Obama administration responded by creating a national commission on forensic science,

00:09:34.120 - 00:09:56.040

which reports to the Attorney General and is coordinated with NIST, the National Institute of Standards and Technology. And this national commission, on which one of our own members, Jim Gates, serves, has been an important forum for discussions of many things, standards for accreditation.

00:09:56.040 - 00:10:16.240

There are discussions going on right now about discovery procedures with regard to forensic science in criminal cases, what information should be turned over to each side, et cetera. There are many important things that are happening there. In addition, something called the Organization of Scientific Area Committees was established

00:10:16.240 - 00:10:41.880

by NIST, where something like 26 different areas have committees in which forensic practitioners primarily are trying to improve best practices across their 26 areas. And continual improvement of best practices is a very important, ongoing thing. We applaud the activity that was launched by the administration.

00:10:41.880 - 00:11:02.400

But in reviewing the area at the request of the president, our job was to ask what's not happening enough. And the one area that we've really focused on is the question of the fundamental definition of what does it mean for a forensic science method, in particular, a pattern comparison

00:11:02.400 - 00:11:21.320

method to be based on reliable principles and methods. The law requires that for evidence to be introduced by an expert in a federal court, it's called Rule 702 of the Federal Rules of Evidence we've come to learn, it must be based on reliable principles and methods.

00:11:21.600 - 00:11:37.480

But the law doesn't define reliable principles and methods for all possible or purported scientific or technological methods. It gives those terms. What we've set out to do in a way to be helpful to the federal courts, to prosecutors, to

00:11:37.480 - 00:11:56.960

defense attorneys, to forensic scientists, is to give a clear, crisp description of what it means for a forensic pattern comparison method to be based on reliable principles and methods. A pattern comparison is a subset of metrology, the science of measuring and comparing things.

00:11:56.960 - 00:12:15.440

And happily in this case, metrological sciences, the scientific community has a crystal clear sense of what it means for something to be reliable. And of course it depends on empirical measurement. You know that something is reliable because you've tried it and it gives the right answer

00:12:15.440 - 00:12:31.640

with high probability. So there's a clear set of guidance there. And one of the things PCAST has set out to do is try to explain very clearly what those standards are in a way that all the parties in the field can use.

00:12:31.640 - 00:12:51.880

They're important because they control admissibility in federal court. Now in addition, what we've tried to do is understand how do those principles apply to some areas, some examples of forensic science methods. Here we really first sought to get the input of the whole community.

00:12:51.880 - 00:13:08.680

So PCAST was unusually vigorous in gaining input. We've held a large number of meetings, probably 20 meetings either here in public session or subgroups of PCAST having phone calls with people or our staff doing deep dives to truly learn the literature.

00:13:08.680 - 00:13:29.600

We took the unusual step of making a call on the web asking anyone to offer to us the their input on the questions of reliability of methods to point us to papers in the literature that would point to the reliability of different kinds of methods to offer us advice in any other way.

00:13:29.600 - 00:13:50.920

And we received 74 replies on the web, I believe, and they point to something like 399 different scientific papers which PCAST has reviewed in a few cases is still working its way and reviewing so that we could learn about this subject thoroughly. And we are enormously grateful to all of the individuals and groups who have taken the

00:13:50.920 - 00:14:05.560

time and trouble to write to us and send us their thoughts. Some of them are very extended and lengthy thoughts. All of these will be made publicly available together with the report, all the references that we've been pointed to will be made publicly available with the report.

00:14:05.560 - 00:14:29.240

We've also reached out to federal judges and put together an advisory group of federal judges who have been able to provide us advice on what the law actually says and also how to be able to bridge the communication divide between scientists and judges and expressing clearly what the scientific definition of reliability means in a way that it would be

00:14:29.240 - 00:14:43.400

valuable to folks in a courtroom. So we've done all that and what we will attempt to do and we're still working our way through it and so I can't report anything definitive here because we're trying to be extremely careful in looking at the literature.

00:14:43.400 - 00:15:08.000

But it's to apply those guidelines about what constitutes reliable principles and methods to a number of different areas as an illustration of the kind of reasoning that's necessary. I think it's not going to shock anybody to say because the National Academy said it loud and clear, there are some forensic science methods that do not yet have foundational

00:15:08.000 - 00:15:20.360

validity. There simply isn't evidence of them. But we're going through carefully, we'll make some decisions about how that is. But it's important to know because in many cases it may be possible to get foundational

00:15:20.360 - 00:15:35.200

validity. Remember, foundational validity doesn't mean perfection, a method doesn't have to be perfect to be foundationally valid, no scientific method is. But it does require that its error rates are known under relevant circumstances.

00:15:35.200 - 00:16:04.120

If you can't actually tell me, well, there's a wonderful quote from a federal case, USVE, that someone pointed out to us about smiles, to say the two smiles match, are we talking about just a general smile or a Mona Lisa smile and if you don't know the details of the matching rule for smiles, you don't know whether that's a meaningful statement or a

00:16:04.120 - 00:16:16.200

meaningless statement. So for a whole lot of different reasons, it's important to try to bridge this communication gap. We are really, really admiring of the work that has been going on so far within the

00:16:16.200 - 00:16:29.920

administration on it. And we really see this key issue that everyone has pointed out to us of the need for the courts to be able to clearly define this admissibility criteria and it's the place for us to apply work.

00:16:29.920 - 00:16:50.720

And our report, I'm hoping that before the next meeting we'll have completed that report and be able to bring it to public discussion either at our very next meeting or if we get it all done perhaps at a phone call meeting, we'll see how it's all progressing, will be a useful contribution to what has been an important ongoing process.

00:16:50.720 - 00:17:11.840

I should really say that the higher the quality forensic evidence that we can have in courtrooms and frankly the higher the bar to admissibility in the long run, the better for everyone. Most importantly for the general public because the right people will be caught and punished for crimes and the wrong people won't be.

00:17:11.840 - 00:17:32.800

We will be able to solve crimes better and there's just so much that can come from it. So a lot of people on this PCAST have worked very hard on this project. Mike McQuaid, Jim Gates of course has been working on the commission, Bill Press, Susan Graham, Dan Shragg, Diana Pankovich has been staffing this effort and so while we're not

00:17:32.800 - 00:17:49.560

done and in a position to go put up recommendations for you today, I did want to give this full report here in public session of the work that's gone on and the framing for it and throw it open to any discussion we might want to have here at this meeting. I'll stop there.

00:17:49.560 - 00:18:11.640

So questions, comments, Maxine. There's a very important report and you're doing well, but if focusing on the federal which is appropriate, so many of these cases are at the state level or local level. If the guidelines are accepted at the federal level, how does they implement it at the state

00:18:11.640 - 00:18:28.280

level when it's the interaction? Well of course in the United States we have 51 jurisdictions to think about at least and then there are sub-jurisdictions of those. It happens in this case that the federal courts operate under the federal rules of evidence,

00:18:28.280 - 00:18:53.560

particularly this thing I referred to rule 702. About half the states have adopted the federal rules of evidence in its exact form or in some modified form and so while precedents in the federal courts don't automatically apply to criminal cases in the states, it seems likely to me that if they follow the federal rules

00:18:53.560 - 00:19:13.400

broadly the advice we're offering is applicable to those states. The other 25 or so states I think follow the older what's called the FRI standard after the case USB FRI. That's different but not so radically different that what we're saying can't be understood

00:19:13.400 - 00:19:33.040

within the FRI context. We're not going to attempt to frame this for each of 51 different jurisdictions of course but I think as framed whether you're under federal rules of evidence and what's called the Supreme Court's Dalbert decision or this earlier FRI standard, I think what we say

00:19:33.040 - 00:19:50.600

should be usable by a judge in any of those settings. The question of whether a technology is foundationally valid is not case specific and it's frankly not law specific. No court really wants to admit a technology where nobody's actually checked whether empirically

00:19:50.600 - 00:20:10.560

it gets you the right answer and so that's really what will drive it so even though we are staying in our lane by providing advice to the federal government because we are a federal advisory committee, I believe that advice is going to be quite usable by states as well.

00:20:10.560 - 00:20:29.440

Jim Gates. Thank you and thank you Eric for giving an overview of this project. I just wanted to commend the working group and the group within PCAST that has been doing this because I've been, as Eric mentioned, I think I've used the analogy on Hadrian's

00:20:29.440 - 00:20:52.280

Wall for a year working, over a year working on the forensics commission and one of the things that I really admire about the effort that's underway here is that we've actually segmented the problem in a way that corresponds to the discussion I have seen on the commission. There is indeed the issue of the foundation of validity of science and that is one problem

00:20:52.280 - 00:21:19.160

that one can see in the way that forensics is presently practiced but there's a separate problem which you are very much aware of that when forensics experts get in the courtroom the way in which they testify about the reliability of the results that are found in the laboratories is also itself problematical and although our report doesn't directly deal with that

00:21:19.160 - 00:21:43.880

issue we do call it out and so that's another area which I think that the community, the forensics community itself under the gentle prodding shall we say of our effort will attend to correct matters so I just wanted to say that it's really confident, I have a lot of confidence that we're moving properly in this.

00:21:43.880 - 00:22:03.680

Thank you. Good. Seeing no other flags up, thank you Eric for that summary. We will now move to our panel on near-earth objects if we could have our four panelists

00:22:03.680 - 00:22:41.800

go to the front of the room and we'll get your name tags up there. The topic of near-earth objects is one that I have long been interested in as has PICAS member Chris Chiba. It is one the president is interested in, his interest was particularly peaked early

00:22:41.800 - 00:23:08.400

in the administration when two smallish near-earth objects passed the earth inside of the orbit of the moon that tends to get your attention and it gets your attention even more when you look at the numbers of near-earth objects on trajectories that might ultimately intersect that of the earth and what could happen if a big one hit.

00:23:08.400 - 00:23:27.360

I sometimes say our task is to prove that we're smarter than the dinosaurs most of which appear to have gone extinct about 65 million years ago as a result of a very large asteroid impact in the Yucatan. So I'm very pleased that we have a terrific panel this morning.

00:23:27.360 - 00:23:48.000

The panelists are Benjamin Alvin Drew, OSTP assistant director for space and aviation security. Lindley Johnson, program executive for near-earth object programs and planetary defense officer in the planetary science division of NASA headquarters, the science mission director

00:23:48.080 - 00:24:09.400

at Bavia Law, research staff member institute for defense analysis, science and technology policy institute and Leviticus A. Lewis, chief of the national response coordination branch of the federal emergency management agency FEMA. Their more detailed bios are in the possession of all of the PCAST members in your folder

00:24:09.400 - 00:24:28.320

so I will not take the time of the panel to recite the members many accomplishments. Let me thank you for being with us this morning and turn it over to you. I would imagine you have probably agreed on an order so please take it away. Thank you Dr. Holdren.

00:24:28.320 - 00:24:42.440

Good morning everyone. I'm Alvin Drew. I'm the co-chair of this committee on detecting and mitigating the impact of Damien here along with Lindley Johnson with a charter of basically trying to prevent us from meeting the same

00:24:42.480 - 00:24:59.680

fate as the dinosaurs from large asteroids or being bedeviled in the meantime by smaller ones along the way. For today's agenda I'll talk about mitigating the initial, that committee was set up initially with the idea of trying to deflect or trying to disperse things that were nearby, couldn't

00:24:59.680 - 00:25:15.240

quite pull it out completely from the problem of keeping custody of these things out there. Bavia Law will talk about is this really a problem? I think one of the big problems is it's a very low probability event but very high impact if it does happen.

00:25:15.240 - 00:25:36.040

Lindley Johnson will talk about his ongoing efforts to hunt for near-earth objects. He's going to have for decades now and Leviticus L.A. Lewis will talk about FEMA's preparedness and post-strike response for near-earth objects. So on from that, so just briefly as Dr. Holdren talked about a couple of objects that came

00:25:36.040 - 00:25:55.800

by the Earth back on Ash Wednesday of 2010. A 125 ton object entered the Earth's atmosphere over the South Pacific, left a big fiery trail of plasma over top of the Isthmus of Central America and finally came to Earth's surface over Central Florida.

00:25:55.800 - 00:26:08.320

There were no casualties. The name of the object was the Spatial Discovery and I was on board that object. So let's find it ironically right now I'm in the part of the solution for dealing with near-earth objects.

00:26:08.320 - 00:26:19.320

It might smite the Earth. I went for a few brief fiery moments. That was part of the problem. So the problem with trying to get rid of near-earth objects, even from hitting the Earth, the

00:26:19.320 - 00:26:35.760

southern ones we intend to have come back to the Earth, is hard to deconflate from the rest of the task of one detecting or dealing with the strike. They all come together. So if I'm going to try and nudge an object away from hitting the Earth or to somehow

00:26:35.760 - 00:26:47.520

disperse it, the first thing I need to know is the information I would have gotten from the asteroid hunters out there, like Lily Johnson had been hunting them, I need to characterize them. And to decide what to do about them, I need to know what would have happened if this thing

00:26:47.520 - 00:26:57.200

does strike the Earth. Is this something that's worth writing out or is it something that we need to do something about? And that comes from the folks at FEMA doing, okay, this is something we really don't want

00:26:57.200 - 00:27:13.800

to have hitting us because the consequences are too onerous to bear or even consider. So we have this whole continuous chain. So if I'm just trying to get rid of these asteroids, what do we do from start to finish? And so we have the detection, locating, characterization part, which is the warning part, giving us the

00:27:13.800 - 00:27:28.160

data so I can go and figure out what to do about them, and Lindy will talk about that. The flexure and dispersal, which is the core of what we're working on for this working group, you have that all fails or something just too big or too close. What do we do here on Earth to prepare for that?

00:27:28.160 - 00:27:36.720

And if something does hit the Earth, how do we go back in and clean it up? And then we'll do that. Underpinning all those things, no matter which phase of this process we're talking about, are two basic things.

00:27:36.720 - 00:27:47.400

One of those is the modeling and simulation tools. So you're talking about things of very spare data sets out there, and you're also looking at things that are very long-distance from Earth, and it's hard to gather the data from them.

00:27:47.400 - 00:28:02.120

And so you need some very intelligent guesswork to deal with the large uncertainty bands around these things and where they're going to hit, what they're going to do. So we need modeling and simulation tools, a lot of science behind that. And also, anything that's going to hit the Earth has global implications, even if it's

00:28:02.120 - 00:28:14.360

a small thing, it gets them down in a place that's very sensitive, so you want to be able to have good communications. So the international cooperation part is going to touch on every one of these, no matter which aspect of the problem you're talking about.

00:28:14.360 - 00:28:28.880

So on for mitigation, here's the things that I think about that keep me up at night is they answer the question, what should and can be done, and when do I need to know it by? There's some very different characteristics between a large object that's long-distance

00:28:28.880 - 00:28:47.680

away from the Earth, and we have decades to respond to it, and one that's smaller, but much closer in, like the one that surprised us back in February of 2013. We had no warning on that one, and they're very different things for that whole chain of custody from the time we find it until the time we have to go out and do something

00:28:47.680 - 00:28:55.840

about it. The parts that we have, you see up there in white or something, I think NASA has a very good hand-long. The parts that we need to work on are the ones that target, engage, and assess, or the ones

00:28:55.840 - 00:29:10.840

we really need to go figure out as part of this preparation. The table on the side there talks about the warning time you have before impact. This talks about the size of the asteroid and the basis scope of the effects. I lifted that from the Linn-Lease reports, and you'll see it again probably in one of

00:29:10.840 - 00:29:25.240

these slides there in greater detail. The part I'm more concerned about is up here, is this optimistic part that says provided we're ready to launch anything that goes out into space. By my survey, I don't think we have anything ready to launch right now, so this part looks

00:29:25.240 - 00:29:36.760

more like this. Let me see if I can get this to build. Anything inside of this side of 2026 increasingly becomes more of a civil defense issue, no matter what the size of the asteroid is.

00:29:36.760 - 00:29:48.800

The charter is here to figure out what parts do we need to have ready for launch in time to make this a realistic chart that we had before, so we're not increasingly panicking like this. Something gets closer and closer to us.

00:29:48.800 - 00:30:25.000

With that, I'll hand you over to Bobby Law. Thank you, Bobby. Yes. As Alvin said, I'll be talking a little bit about what is the problem we are trying to

00:30:25.000 - 00:30:38.440

solve here. I'll be quick so we can move on to the asteroid hunting and the really cool parts. It goes without saying that the asteroid threat didn't start with Hollywood making Armageddon a deep impact.

00:30:38.440 - 00:31:01.440

Earth has been bombarded with nearest objects or asteroids and comets since it was formed, four and a half billion years ago. One of the very common examples that we do talk about is the 65 million year ago, Dr. Holdren mentioned the potential comets that may have caused the extinction of not just

00:31:01.520 - 00:31:23.920

dinosaurs, but 75% of all non-avian life on earth. It was about a 10 kilometer wide nearest object. Just to make sure that we don't think this is something that happens in prehistory. When I was starting my first job, I remember watching pictures of a comet shoemaker, Levy

00:31:23.920 - 00:31:40.800

9, hit Jupiter. It had 21 pieces, some of them as large as 2 kilometers in diameter. It was moving extraordinarily fast, 60 kilometers a second. It hit Jupiter with the impact of the entire arsenal of earth's weapons.

00:31:41.680 - 00:32:06.360

This was the kind of hit that would take earth out, essentially. Back on earth, 50,000 years ago, there was an impact of a near-earth object, a neo. We think it was about 45 meters wide because it was metallic and not stony. It hit the ground, stony, neos tend to be airbursts or get destroyed in a fireball in

00:32:07.360 - 00:32:22.360

the atmosphere. Again, moving pretty fast, 20 kilometers a second. The blast equivalent was about two and a half megatons of TNT equivalent, and it made a crater that was about a mile wide and five and 70 feet deep.

00:32:22.360 - 00:32:44.360

A small object 45 meters, making a pretty big impact. More recently, in 1908, in Tunguska, Russia, a fast-moving neo. This was most likely stony. Fragmented at about 10 kilometers above the earth with the blast equivalent of five megatons

00:32:44.360 - 00:33:05.360

of TNT, and it destroyed about 2,000 square kilometers of forest. Just very, very recently in 2013, in Chelyabinsk, Russia, there was an airburst, very small sized, a neo as compared with the others. I talked about 20 meters, but it had the equivalent of 30 Hiroshima bombs.

00:33:05.360 - 00:33:19.360

Again, very powerful. Chelyabinsk is a small city. It injured about 1,600 people and caused about a billion rubles in property damage. Just giving you a sense of...

00:33:19.360 - 00:33:34.360

We have the very big, the 10 kilometer size asteroids, and then we have the tens of thousands of meter rights that enter earth's atmosphere every day. There are dust size, actually probably more than tens of thousands. What is the right size to be worrying about?

00:33:34.360 - 00:33:58.360

I think that's an answer that we don't really know, and I think Lindley's going to talk about that threshold. Just to give you a sense, if there was a Tunguska type neo coming over, let's say, New York or San Francisco or Washington, D.C., this is going to be a pretty major impact. New York alone has 8 million people, over 8 million people.

00:33:58.360 - 00:34:17.360

An airburst might have... Tunguska had 2,000 square miles of trees that were flattened. We're talking about flattening a major U.S. city. I don't want to find a silver lining here, but if we were to, this is a local issue.

00:34:17.360 - 00:34:43.360

There is likely going to be no global challenges like kicking up enough dust that we have a nuclear winter type of situation with something this small. Thinking again about smaller neos, there's a lot of stuff going on on this chart, but the thing I wanted to point, if the x-axis is the equivalent of size, how big the neo would be and the y-axis is the number.

00:34:43.360 - 00:35:18.360

As you look on the right side, you see that the kilometer and on, there is a thousand-ish neos that are above that range, but if you start to look at the smaller Tunguska sizes, you see that we are looking at about a million asteroids just in that size. It's a substantial number. By extension, the smaller entities tend to come more frequently.

00:35:18.360 - 00:35:43.360

Again, looking at the chart going from the bottom to the top, for the very large objects, 10 kilometers, we think there's about four of them, and they come every 100 million years or so. Not that it's not something to worry about, but potentially not to worry about in the very short term. Moving up, as you come to the 50-meter asteroid, it comes about every 2,000 years.

00:35:43.360 - 00:36:12.360

Again, a lot of these numbers are approximate because there's only so much research we have, and we actually don't know how many there are. Moving smaller sizes, we are looking at impacts that maybe every 100 years or so, and there's more than a million of those. Of course, the important thing is that we actually don't know where they are. Going back to the chart I had before with a little bit more information on it, you see that the red line is essentially the numbers following the power law.

00:36:12.360 - 00:36:42.360

The blue line is how much of the survey has been complete, and Lindley will talk more about that, but the green line, the green bars are the important things to see. You see that up until about a kilometer, our survey is pretty complete. We know what we've detected them, we know where they are, but as you start to go leftwards, you start to see that our inventory is really incomplete. Obviously, we need to worry about larger asteroids, which are the civilization enders, but we need to worry about the smaller ones too.

00:36:42.360 - 00:37:03.360

Not only because they can cause damage, that could be substantial, but they just come more frequently. In our lifetime, we could see one of those. I think our saving crisis is that the population isn't everywhere. Most of Earth is water, and even when there's land, there's different levels of population distribution.

00:37:03.360 - 00:37:38.360

More likely than not, we hope that a small near will hit where there's no people. However, hope is not a plan, as we've heard. To talk about the plan, I would like to turn it over to Lindley Johnson from NASA. Okay, now that Bavio is scared you have to do it, I'm here to kind of tell you what we are doing at NASA and with our partnerships across the U.S. government and internationally in this area.

00:37:39.360 - 00:38:17.360

I head up what we have come to call the Planetary Defense Coordination Office at NASA. This office was recently set up, formalized, but NASA has, as you will see later in my briefing, has been looking at this problem for a number of years. But our office is set up to lead both national and international efforts, first of all, to detect any potential impact threat. You've got to find them first, you can't do anything about them unless you can find them, and hopefully a number of years before they are impact threats.

00:38:17.360 - 00:38:53.360

So we have a chance to do something about them. We also are involved in a lot of work to figure out exactly what is the threat and at what size objects and what composition of objects is. Asteroids, of course, come in different compositions as Bavio alluded to, whether they are metal rich or whether they are simply fluff balls of a lot of volatiles like comets. So those same size objects with those different compositions could do different levels of damage.

00:38:53.360 - 00:39:48.360

So some of our work is to assess four different size objects, what the threat level might be. And really to establish at what range threshold are the objects large enough that we for sure want to find them far enough ahead of time that we can do a space intercept to deflect them off the impact trajectory. And below what size level might we be able to be able to take the hit, particularly if we know that the object is headed to a mid-ocean or largely unpopulated area. You know, it's probably not worth the cost of undertaking a space mitigation campaign if we know it's only a 30 or 40 meter size object that's going to hit in the middle of the Pacific.

00:39:48.360 - 00:40:29.360

Assuming that it doesn't cause a large tsunami and that's something else we are studying in these potential effects areas to understand what an impact cause tsunami might amount to. There's a lot of dichotomy right now in the modeling for impact cause tsunamis and we're working that out. And then also developing strategies to be able to mitigate impact effects both in space and here on earth working with my colleague here over on the left. So the Defense Coordination Office has set up this way.

00:40:29.360 - 00:41:25.360

First of all we have the New York object observation program. This is a legacy program that NASA has been doing since 98 and we have another number of projects that are searching for these objects as well as taking data to characterize them so we understand what the population is out there and what it amounts to. Then down the middle here we have a office that is working, officers that are working, the interagency and emergency response, working with FEMA and others as to first of all when we find an object that is an impact threat getting the warning out to the right government agencies and others and then working with FEMA in emergency response planning that work goes on right now and LA will tell you more about that.

00:41:25.360 - 00:42:22.360

And making sure that we are prepared as much as we can be to be able to respond should an impact threat be detected. And then in the third area we are working mitigation research projects trying to figure out if we are able to find them far enough ahead of time what we could do in space to deflect these objects off the course or at least if it is a more short term warning what could we do to break them up so that the impact on the earth would not be as bad and perhaps our atmosphere can take care of what is left as it does every day for us with the small stuff we get hit every day with. So the observation program started in 98 with agreement with Congress that we would find at least 90% of the one kilometer and larger objects.

00:42:22.360 - 00:43:29.360

We were able to reach that goal in 2010 having found some 850, 800 or 70 of these objects in a population that is a little shy of a thousand. That in itself taught us something about the population because going into this we thought the population was probably about twice that much but as we conducted the survey we found that it was only about a thousand one kilometer larger. But in the meantime our objectives were moved as we began to realize that objects a lot smaller than one kilometer in size could be a bad day for an area of the earth. And with the Authorization Act of 2005 Congress amended NASA's charter to make this work one of the important activities that NASA should be involved in and established our current objective of finding at least 90% of the 140 meter and larger objects within a 15 year period.

00:43:30.360 - 00:44:14.360

This is a simplified version of the chart Babio showed. This is what the statistical analysis of the known population shows us that the entire population for the various sizes is. This is work that's done by Al Harris who is a very prominent in the field of working the NEO population and understanding the population. We update our population model every two or three years. The latest update was done in 2004. And Babio showed you a version of this chart which shows that for the one kilometer and larger we have done a good job of finding most of that population.

00:44:14.360 - 00:45:01.360

We feel we know of any objects that are five to ten kilometers in size. In larger we know where all those are. The one kilometer to five kilometers. So there's still a few out there to be found in the order of a few dozen. But our original objective has been accomplished. And so now we're working on this objective of finding them all down to 140 meters or so in size. And as you see with the population completeness curve there in the blue that as we drop below one kilometer or so in size our understanding of the population drops off rapidly because the current assets that we use simply weren't designed or weren't designed to find asteroids in the first place.

00:45:01.360 - 00:45:51.360

The assets that we were using we've adapted to it. But they certainly aren't highly capable of finding the smaller ones in a timely manner. These are current search programs that are funded by NASA. But first of all I want to point out the two very important data analysis and processing nodes in our system. One is the Minor Planet Center. Minor Planet Center has been in existence for decades. Their job has been to catalog the smaller bodies in the solar system. Asteroids, comets, hyper belt objects. And they do that for us. They are the International Astronomical Union sanctioned database.

00:45:51.360 - 00:46:36.360

All astronomers that do this work in the world know to send their data into the Minor Planet Center. And you can see all of their data there on their website. And we also have established the Center for NEO studies out at the Jet Propulsion Laboratory. That is where the more precision orbit determination is done. Once Minor Planet Center is established what the orbit of an object is determining that it's a near-earth object for instance, then JPL takes over and does a more high precision orbit determination with all the observation data that we have. And they are able to project confidently the orbits out to about a century to determine if an object is an impact hazard to the Earth.

00:46:36.360 - 00:47:22.360

Not only the Earth but also other planets, large bodies in the solar system, were to be predicted impact on Mars. We want to know about that. That would be a very interesting science experiment to see an impact on Mars. And that's all done through a heavily automated system. It only involves human interaction with it when an object pops up that, you know, does show high probability of impact. And they go in and make sure that we really understand what we're looking at. I should also point out that we do this in parallel with our colleagues in Europe.

00:47:22.360 - 00:48:09.360

They have a similar system to JPL in predicting the hazard and the potential of impact and we compare results, of course, before we would announce any particularly impact. The three telescope systems you see along the bottom, these are some of the, doing most of the work in discovering asteroids. Linear is a project that's operated by MIT Lincoln Labs using the Air Force Space Surveillance Assets. They were using the one-meter ground-based electro-optical deep space surveillance system for a number of years, prototype telescopes of that.

00:48:09.360 - 00:48:53.360

But as the Air Force has moved to this new Space Surveillance Telescope that's been developed by DARPA, MIT Lincoln Labs is working with us and the Air Force to install asteroid detection and tracking capability on that telescope as well. Right now it operates down in the White Sands. It's due to move to Australia here in the future. University of Arizona operates three telescopes for us, collectively called the Catalina Sky Survey. And the University of Hawaii operates the Pan Starter System 1.8-meter telescope and a second 1.8-meter telescope just now coming online this fall.

00:48:53.360 - 00:49:39.360

We've also adapted one space-based asset. The White Field Infrared Survey Explorer was an astrophysics mission developed to build up an infrared background of the sky, a map of the sky for James Webb Telescope. We quickly, as we were developing it, realized that with all the images it was taking of the infrared sky, we'd find asteroids in there too. And so we adapted it with additional processing, a project called Neowise, to do that during its prime mission in 2010. But as our program has gotten more funding here in the last few years, we were able to bring that. It only had a one-year mission with astrophysics and then it was put into hibernation.

00:49:39.360 - 00:50:37.360

But as we got additional funding, we were able to bring that satellite out of hibernation and put it in full-time use as an asteroid detection and tracking. So this chart shows you the discoveries each year of the Near Earth Objects by these various projects. You see Linear and its earlier version was one of the big discoveries through the early part of the program that started in 98. Then Catalina Sky Survey became the prominent project and now PanStars is picking up a lot of the objects in the current endeavor. With PanStars coming on full board, we've been able to increase our discovery rate per year up to about 1,500 Near Earth Objects discovered each year.

00:50:37.360 - 00:51:29.360

In 2016, we're well on our way to breaking through that record even with already 823 found in the first five months of this year. This is a cumulative total of NEOs in Near Earth Asteroids, I should say, that have been found since at the start of the NEO program in 98, there were only about 500 known Near Earth Asteroids. And as our capabilities have increased, our discovery curve is going exponential there, as you can see. So total of all known Near Earth Asteroids, a little over 14,000, we'll break through 15,000 before the end of the year.

00:51:29.360 - 00:52:07.360

Also, there's 106 comets that are in orbits that come within 30 million miles of the Earth too, and we certainly want to know about those as well. Almost 1,700 of them are in orbits that we consider hazardous, i.e. they come in within 5 million miles of the Earth's orbit. Therefore, they could over time wonder into collision with the Earth. Of the one kilometer in larger, we have 875 in our catalog of which 157 are in the hazardous orbits.

00:52:07.360 - 00:52:45.360

For the objective, 140 meter in larger, a little over 7,000 now in a catalog, about half of the population are larger. But that is an observational bias. The smaller objects are certainly a lot more of them than there are of the 140 meter in larger. It's just that we've only been able to develop the capability to see the smaller ones further out in the later years. The earlier graph is a little deceptive. The small triangle looks smaller than the big one kilometer in larger triangle.

00:52:45.360 - 00:53:27.360

So I break it down here with a pie chart. 140 meter population is estimated to be between 25 and 26,000 objects. The one kilometer in larger represent a little over 4% of that population of which we have found most of those 300 meters to one kilometer in size. It is about 30% of the overall population and we have found about half of that population 300 meters to one kilometer in size. But once we drop down below that, the larger part of the population, 140 meters to 300 meters in size, represents about 65, 66%.

00:53:28.360 - 00:54:00.360

And we have yet to find 60% of that overall population. So together, the two unknown areas is we still have about 75% of this population to find in our current objective. And we won't find them all by 2020. We use a number of assets also to characterize the objects once we find them, more detailed observation of them.

00:54:00.360 - 00:54:41.360

We use the two planetary radars that we have, the Goldstone as part of NASA's Deep Space Network, and then the Aerosceval, a laboratory that's owned and operated by the NSF that NASA has the radar capability on. NASA has its own infrared telescope facility, ground-based 3-meter telescope on a Mauna Kea that's used a lot to determine the composition of these objects once we find them. And we've also pressed into service the Spitzer Space Telescope where we can where it has access to these objects to learn more about their sizes.

00:54:41.360 - 00:55:18.360

So we contribute to the continued operations of Spitzer. Just a few words about international cooperation. In February of 2013, when Shalu Bin's happened, I and my colleagues from other nations were actually at the Committee on Peaceful Uses of Outer Space, a scientific and technical subcommittee meeting about to brief our recommendations as to what the international community ought to do about near-Earth objects.

00:55:18.360 - 00:55:55.360

And lo and behold, that morning is when Shalu Bin's happened. So that was just Mother Nature putting an explanation point on our recommendation. What's been agreed to in that forum is that the member states should operate as part of an international asteroid-warding network, pool our assets of observers, analysts, and modelers to find, detect, track, and characterize the population to the best of our collective ability.

00:55:55.360 - 00:56:39.360

And then also established an international forum for the space agencies to be able to gather and talk about technologies and techniques that are needed for an in-space mitigation, part of what we call the Space Mission Planning Advisory Group, or SAME PAGE, so that we're all working together to develop what the best response might be to a detected impact threat. And we do that all with the committee and with the support and sponsorship of the Office of Outer Spaces Affairs. International Asteroid Warning Network in 2015.

00:56:39.360 - 00:57:04.360

These are red dots that depict all of the observatories around the world that contributed observations to the Minor Planet Center in 2015, some 20 million observations from almost 40 countries. You saw this chart earlier. This comes out of the report by the National Research Council study that was done in 2009, reported out in 2010.

00:57:04.360 - 00:57:32.360

Of what might we do to mitigate an object deflected that was on course for an impact. The leading three technologies, techniques that are thought to be most viable for doing this, given our current technology, is of course always a nuclear option, which we'd prefer not to use. We'd like to find them much further out that we could do something else.

00:57:32.360 - 00:58:05.360

But if they are particularly large, that probably is our only recourse, have pack enough energy to do what would need to be done. But there are two other techniques which would be effective against the larger part of the population, 100 meter or so in size. First being the kinetic impactor. You just hit the object with spacecraft going fast enough that it would impart a change in velocity.

00:58:05.360 - 00:58:53.360

And if you do this several years in advance, you only need to change the velocity of the object by a kilometer, I'm sorry, a centimeter or so per second. So that a few years down the line, instead of it being impacted, it'll be a complete miss of the earth. Then if you've got more time, this idea of a gravity tractor, which is very elegant in the fact that you don't have to even touch the object. You just use the laws of gravity to position the spacecraft off to the side station keep by the object and use gravity as a tug rope to pull it off onto another trajectory over a course of time.

00:58:53.360 - 00:59:48.360

This chart may be a little conservative on the area in which the gravity tractor might be effective. But certainly if you've got 15 to 20 years and it's a small object, this is a viable technique and we are working to demonstrate these two techniques, kinetic impactor and gravity tractor. Also working in a nuclear realm though, NASA works with the Department of Energy and the National Nuclear Security Administration in investigating the approaches that might have to be taken either in a short term warning or with a large object. As far as the kinetic impactor technique, we are working with the European Space Agency on a concept called the asteroid impact deflection assessment.

00:59:48.360 - 01:00:33.360

We are currently in formulation phase A studies where the European Space Agency will launch the asteroid impact monitor. That's the spacecraft to go out and rendezvous with the potential target and understand this composition and provide more detailed data on the object than we can get from the ground. Then the U.S. NASA would come along with the DART, the double asteroid redirect test. The reason it's called the double asteroid redirect test is that the selected target for demonstration of this technology is the asteroid ditimos which is a binary near earth asteroid.

01:00:33.360 - 01:01:09.360

What we would do is hit the moon of ditimos, affectionately called ditty moon. It's not an official title for it, but the reason we want to demonstrate this on the moon is one, it's about the canonical 140 meters in size that we are searching for. We want to make sure we can hit a target that's that small, millions of miles in space. It's also easy to see the effects of the deflection and that the period of the moon would be affected.

01:01:09.360 - 01:01:44.360

That's an easy thing to be able to see even from ground based telescopes. This would be a very important demonstration of the kinetic impactor deflection. The other nice thing about it is it's an international collaborative effort, not only with the Europeans, but the Japanese are also interested in participating in this mission. And then the other large project that we are working with, the coordination as part of our office,

01:01:44.360 - 01:02:35.360

and coordinating with the asteroid redirect mission, which is a leading program for NASA to demonstrate the technologies and techniques that are needed on our journey to Mars, being able to move large masses in cargo from the Earth to Mars to support the human exploration of Mars. The Near Earth Object Program provides the identify segment of this mission using the technologies and capabilities that we have to detect and characterize Near Earth Objects. So we have a few valid candidates for this mission already, but we continue to search for even better candidates.

01:02:35.360 - 01:03:28.360

We still have a few years before the decision would have to be made as to what the target would be. The asteroid robotic redirect mission is depicted here in the middle where the robotic spacecraft with the solar electric propulsion technology and the grappling technology that's being developed would go to a Near Earth asteroid, collect a large tens of ton boulder off the surface, and grasp that in its arms and bring it back to Cessilunar space where the human crew could go up and explore, and take samples and practice in deep space, I should say, docking in rendezvous and EVA techniques.

01:03:28.360 - 01:04:11.360

As far as a robotic spacecraft, we are particularly interested in it from the planetary defense aspect, going to this Near Earth asteroid, and it's mass off the surface because before it comes back to the moon, we want to do a planetary defense demo of the gravity technique with this mass, station keeping off to the side of the asteroid to slowly pull it off of its trajectory to demonstrate that that technique really does work and how effective it can be. So with that, I'm going to turn it over to LA.

01:04:11.360 - 01:04:59.360

Good morning. Thank you. Thank you for inviting me to this. I never thought I'd be here with an astronaut and two rocket scientists being a FEMA guy, so it's pretty neat for me. So again, I'm LA Lewis. I'm from FEMA. We've been working together. I've been working with Lindy since 2010 with the letter that was sent to Congress which talked about what the responsibilities were. The initial tasker was to come up with a warning mechanism so this event happened, and Lindy and I working together. FEMA already has an in-place national warning system that we've used for other emergencies, and we're going to probably work on modifying that existing framework for just scenario.

01:04:59.360 - 01:05:46.360

So we formed the Planetary Impact Emergency Response Working Group back in 2013. It's a joint effort sponsored by the FEMA Response Directorate and the Planetary Science Division at NASA. Our mission at the Peer Week is what we call it, is to provide a forum for developing essential information and recommendations for our senior leadership so they can make informed decisions to respond to what we consider is going to be a rather unique type of emergency. Everybody's used to hurricanes, tornadoes, earthquakes, etc., etc. And their earth object impact, I don't think it's been witnessed by anyone of a large size since Augusta, I guess.

01:05:46.360 - 01:06:31.360

So IntelliBenz was a good wake-up call. In fact, I got a lot of phone calls from my colleagues wondering why we were not warned about the IntelliBenz object, so that got a lot of interest there. So that's what the Peer Week does. Our role, basically, is to extend working on what will we tell the emergency managers in the United States? How do we send this information to the public? What do we tell them to prepare for? How do we give emergency managers that right information so that they can respond to this and their localities and their states? Again, everybody's used to seeing a hurricane cone on the news. No one will be used to seeing Monte Carlo points on the map when we project asteroid impact.

01:06:31.360 - 01:07:11.360

So we're going to have to be able to explain that and also make sure that we are considered the primary source of knowledge, as you can imagine. Once this gets out, it won't be a secret that there's going to be an impact and all kind of experts will be on the news and will be competing with them to make sure we get the right information. So right now, it's going to be analogous to re-entering large space debris and hurricane warning procedures. We have that down. We know how to do that. We have procedures in place whenever we have re-entries of large satellites, for instance, the ISAT and other things that have come back to Earth.

01:07:11.360 - 01:07:59.360

Trim a couple of years ago, we went through a whole procedure on that. So the emergency community is used to that. So what we're going to probably do is work on, use that as a basis for moving forward. And the other unique challenge of this, as we had our peer week meeting a few weeks ago, the timelines for decisions. If you have hours or days or weeks, it's one way you have to alert the public. But if you have years or tens of years, how you do that to the public and prepare emergency managers is another story. People are not going to want to pay attention. And I think the scenario that we practice on our tabletop exercises is that Americans will first get excited about it.

01:07:59.360 - 01:08:45.360

Maybe when we first announce it, if you say it's going to be an impact in 10 years, nobody will pay attention to it. And then in a team of six months, everybody will ask us, okay, what's the plan and what are we supposed to be doing? So we recognize that to be a difference. And of course, Linda's explained what NASA's role is. NASA is the lead in this. We'll be depending on them for the scientific information and the information on where it is popular going to impact. And also the efforts that they're making in modeling. It's one of the things we want to do also in informing emergency managers and leadership and responses. Why it makes a difference as to how this, why is it important to know what the asteroid is made of, whether it's a stony body or whether it's iron, nickel or metallic and how large it is, how fast it is.

01:08:45.360 - 01:09:38.360

That all makes a difference as the impact on the ground, how much damage there's going to be. And that helps us in our preparation efforts. So again, the period we go is to actually gather all of this information. We kind of act as a clearinghouse for our senior leadership, particularly the emergency management community, and translate all the science and modeling data into real actionable things for the emergency management agencies to do. So how long do they have? Is this different from any other response we have to make? One of the things we're working on, for instance, on Lindsay mentioned tsunamis. You know about earthquake-based tsunamis? We're assuming tsunamis will be the same, but if it's an impact tsunami, will it be the same? You don't know that. We're still working on that. Will it affect when it comes on land?

01:09:38.360 - 01:10:45.360

Will it be different depending on how far out in the ocean it is or how close it is on a continental shelf? Is the west coast different from the east coast if it lands off the ocean? So those kind of things we need to know. Again, the membership is all of the, we have two kind of groups that we work with, interagency at FEMA. We have the emergency support function leadership group. There are 15 emergency support functions. There are 12 recovery support functions. The two different parts, emergency support functions, are the guys who are going to respond, who will help aid the federal coordination. Recovery is what's afterwards. So say we have a large impactor, and it's going to impact a significant part of U.S. infrastructure or a major U.S. city. You know, what are going to be the long-term goals for recovering from that? And will that be different from a devastating earthquake or whatever, depending on the size of the impact? So we want to be able to provide those kind of groups the specific information that's unique to an asteroid impact.

01:10:46.360 - 01:12:12.360

Again, I won't read all of this. This gives more of the details. You can read it at your leisure. Just kind of telling you that the period week is going to try to coordinate all of the different activities that can happen. The main thing is emphasizing what's different about an asteroid impact as opposed to an earthquake or a hurricane or some other emergency. And this is not to say that we're spending nights staying awake at this. This is another emergency on the list of emergencies that we may have to respond to. Because of the unique nature of it, we're starting our partnership with NASA to actually make sure that we understand this phenomenon, understand the science behind it, why the science is different from one of the other emergencies that we're familiar with, and to make sure that we relay that to our senior leadership. We're going to have teams. Right now we have two teams. With the composition of the period week, we'll divide it up into teams where we have specific things. One of the two projects that we're working on now is trying to narrow down what should alert message out to the public, say, depending on the scenario. Whether we have days, hours, or if we have, we also understand there is a possibility of no notice impact. And then we'll handle it like our other emergencies. We're not planning on developing a specific plan for asteroids at this time.

01:12:12.360 - 01:13:31.360

However, we do have federal interagency operational plans which address the various types of emergencies. So depending on our information gathered in this effort, we'll be using those already determined responses to kind of model what we'll need, in particular for an asteroid impact. Our next steps, we're going to work on, again, the information for warning, make sure we are tied into what the international asteroid warning network is going to do, and whether we're going to probably use variations of our current satellite reentry message format, but depending on how the research is going, we may have to modify that somewhat. And our next big thing right now is planning for our third tabletop exercise. I'm going to try to take a slightly different approach this time. We've done two tabletops here in Washington, D.C. We like to think about getting out of Washington, D.C. to get another flavor. And then the next tabletop actually invite the city that we have chosen for our destruction, invite their emergency managers and state emergency managers, and actually work around, actually work through the exercise with them.

01:13:31.360 - 01:14:28.360

And then that way we'll get a different flavor from just having the Washington experts and thinking that we have a good idea of what they're going to ask, but it'd be good to see if we have some local officials in the next exercise to get their reaction to this event. And that's essentially our part. We hopefully won't be needing the services of our colleagues here will be successful and they won't be needing this. Thank you. Good. We have time for a few questions and there are probably more flags up than we might really have time for. So I urge everybody to be concise. Chris Chiba is first. Thanks for a fascinating set of presentations and I can't tell you how pleased I am to hear that all this planning is going on. It's really good news.

01:14:29.360 - 01:15:28.360

As you as you will know, having decades of warning time would make a tremendous difference, both across the board, huge advantages, including the opportunity to try more than once if we don't quite do it right the first time. My question comes in two parts. I'm sorry, John. My first question is, given that given the I would say almost overwhelming importance of having warning time, what is in fact now your estimate given the ongoing search searches for objects larger than 140 meters? What is your estimate for when you'll have a catalog that's more than say 90% complete. And then in the event that we do find an object that's going to hit us some decades down the road, not only do we have to figure out how to deflect it, but I'm curious what you're thinking is about what kind of dialogue takes place within the international community about whether to deflect it, how it's deflected, who has responsibility for making that decision.

01:15:29.360 - 01:17:05.360

Okay, well, with the pace of the current assets that we have searching, as I said, we're finding hopefully this year we'll be up to about 600 to the 140 meter and larger objects, but we have about 18,000 yet to find. And so we kind of do the math there. You know, it's about 30 years at the current at the current pace. We do have plans for additional more capable assets coming online. For instance, we are working with the National Science Foundation and the large synoptic survey telescope of integrating its capabilities into our network as it comes online and working with them to develop the right techniques for using the vast amount of imaging that they're going to be taking to detect and track asteroids in that large data set. But the really most capable way of finding these objects would be to be able to do it in the IR. And to do that, you've got to get above the Earth's atmosphere. And so a space-based IR spacecraft that is specifically designed to detect and track asteroids would be the way we'd like to be able to go.

01:17:06.360 - 01:17:51.360

And so we've had the resources to do that. To your other question and working with the international community, that's what SamePage is all about. It is to try to put at least a template into place with the international community of how we would work together from the warning that we have an impact to putting an international campaign together to go do something about it. Disrupt it depending on what kind of time we have available and what technology we have to do that.

01:17:51.360 - 01:18:18.360

It's not just the technical side of it, but we are also looking at the political issues and the international legal issues as well in doing this as a part of that forum. Jim Gates is next. Thank you, John, and thanks to the panel for that excellent briefing. My question has to do mostly with LA's presentation at the end.

01:18:18.360 - 01:19:00.360

You talked about the planning for disaster preparedness of the nature that a tornado, I'm sorry, hurricane or re-entering a space vehicle. I'm a little bit worried about that because you see, I can imagine a scenario where if we have sufficient warning, non-experts get into the discussion outside of the government. And in particular, I'm wondering about what is your thoughts about external media validators who would then reinforce your message and actually get the proper information out. Because in this age of loss of authority because of the internet, there are going to be a lot of people out there putting out a lot of misinformation.

01:19:00.360 - 01:19:17.360

And I would like to hear something that you're thinking about how to deal with that. The idea of a civilization extincter, you can imagine someone saying, gee, this is the end. And it may not be the end. So what are you doing to deal with that problem?

01:19:17.360 - 01:19:34.360

Thank you, sir. That's a good question. And both of our table tops previously, we've actually discussed this possibility and it took up quite a bit of time. We also talked about it at the last two planetary defense conferences.

01:19:34.360 - 01:20:02.360

We were talking about the idea that because of the availability of this information and internet, that unlike the movies, there's no way they're going to build a spaceship in orbit for a year. Nobody sees it. As soon as it's out, we give ourselves maybe 30 minutes before it's out that somebody knows that there's going to be an impact and it's all over. And we have actively discussed how do you counter all the new experts that will show up on CNN and Fox and MSNBC.

01:20:02.360 - 01:20:53.360

So what we're thinking is that when we had the meetings with the International Asteroid Warning Network, we got involved in the media. But we also had to make sure that we're as upfront as possible with the public so that we get out in front early to show that this is the authoritative information and not making it so dire that people think there are no options. What we should do, whether we should have a website set up that people can go to and look at the data themselves, have people working with our public affairs, our external affairs personnel and explaining the phenomenon to people right away, explaining what asteroids are, where they come from, but it's trying to get out as much information upfront as soon as possible.

01:20:53.360 - 01:21:05.360

But we recognize that there will be competition from the other experts. I know my other panelists have thought about that as well. And what about religious leaders? I mean, because this will become such an issue. Yes, sir.

01:21:05.360 - 01:21:36.360

Of course, the other part is this, it means we can do a deal that is to bring down some uncertainty bands around the data. One of my fellow astronauts and Asteroid Hunters, Ed Lu, he used to put an exercise where it's a near miss to see what the effect something with a big uncertainty would have on just property and property insurance values if something came by as everything that would be under the potential footprint becomes uniquely uninsurable for a brief period of time, what that does to markets. And so the best you can do for our inputs, and I can't do much about the culture part, is we can get those numbers as tight as we can to go, yes, it's going to be close, but yes, it's going to miss,

01:21:36.360 - 01:21:56.360

or this is the circle in which it's going to hit. And the more uncertain we can knock off of it, the less speculation that you leave out there for the Johnny come lately. Susan Graham. I too want to thank you for these presentations, which are very interesting.

01:21:56.360 - 01:22:41.360

I'd like to know more about the international aspects of these activities. You touched on some of it in response to Chris, but it would seem that the first stage you described the developing the inventory of objects, the detection and so on would be an issue in which there's international interest. There's no reason why it affects us more than it affects any other country. And yet your description was mostly about what we're doing to do this detection, which gets increasingly expensive as we get to the harder things to detect.

01:22:41.360 - 01:23:18.360

So what's going on internationally to share this responsibility as well as sharing the information? Well, it is true that the US does have the largest effort here. The US systems are finding over 95% of the objects every year, and we have the most capable systems. So that was part of the idea about establishing the International Asteroid Warning Network with the endorsement of the United Nations.

01:23:18.360 - 01:24:02.360

So that other countries, first of all, you know, could learn about what's being done and what can be done and how they might contribute. One of the things, in fact, just this year at the committee meeting, we had an open forum with the international community describing to them about how we do what we do and how observatories around the world could join the effort and become part of the network, how to send their observations into the Minor Planet Center and those kinds of things. One is it's informing them about how they can participate and then continuing to encourage them to join in the effort.

01:24:02.360 - 01:24:27.360

I could also add, so there's, I mean, I think we did talk more about the US efforts, but it is an international community. The last Planetary Defense Conference was in Italy, and it was very European heavy, and there was Asian participation. I mean, maybe not as much from China and India, but, you know, enough. And the next one is in Tokyo. So there is international interest.

01:24:27.360 - 01:24:52.360

It's just that I think for the moment, most of the detection efforts tend to be US-centric, but I think on the deflection, or at least on the mitigation piece, there is more effort. ESA has some programs, Neo Shield is an ESA program, right? So there is European and Asian programs on the topic. Yeah, well, they actually came to us with this idea of the asteroid impact deflection assessment mission.

01:24:52.360 - 01:25:14.360

That's, you know, largely a European-initiated mission, and we are becoming a part of it if they go forward with their spacecraft. So our intent is to go forward with ours. Bill Press. Thanks.

01:25:14.360 - 01:25:33.360

So I guess my question will sound a little bit like a joke, but I don't mean it that way. My question is about the lawyers. What's the legal framework for this? Can the President, as Commander-in-Chief, order an asteroid to be deflected? Does it take congressional action or international action?

01:25:33.360 - 01:26:03.360

What's the relevant body of law on this? Well, that is one of the issues and areas that St. Page is dealing with. In fact, one of the sub-working groups of St. Page is a legal group, which we're establishing this year. St. Page has been going for a little over a year now, but there are a number of legal issues with this.

01:26:03.360 - 01:26:41.360

The space legal community has discussed and talked about this for a number of years, but not under, you know, a discipline form, I guess I would say. And so we're trying to do that under the same page. He has also said existing laws, you know, that one of them prohibits nuclear weapons in space, so one is a nuclear warhead, not a weapon. The other part is, you know, the part about extracting resources for our own use, because I was jokingly, you know, suggested that one of the ways to get rid of our bigger asteroids

01:26:41.360 - 01:27:09.360

is to clear it open for exploitation and have people out there and mine it into extinction. Of course, there's laws about that, but what you can bring back and declare sovereign. Yeah, well, you know, and one of the questions is if the United States or the United States and the Europeans attempt a deflection mission and it doesn't quite succeed, you know, and we move an asteroid that was going to impact in the middle of Pacific to where it's going to impact somewhere in Southeast Asia, who's responsible?

01:27:09.360 - 01:27:46.360

Or on a smaller scale, if you break it into small pieces and one of those pieces hits me, who do I sue? Unless it's a very small piece. All right, I'm going to have the last question here because we have pretty much exhausted the time, but I have both one comment and one question. The comment is, I think all of us in the science and engineering community need to get better at communicating the nature of stochastic processes.

01:27:46.360 - 01:28:17.360

And the reason is when you tell somebody that the mean return time for an asteroid of this size is a million years, most members of the public think that means this is only a long term issue. We don't have to worry about it in the short term because it's probably going to be a million years till it happens. I think all of us in this room probably know that an annual probability of one in a million doesn't mean you can safely wait a million years. This could happen tomorrow. It could happen in five years or ten.

01:28:17.360 - 01:29:01.360

These probabilities are mean return times, not predictions of how long it's going to take to come back. And I think this is a big problem because it has led people, I believe, including the folks controlling the purse strings, to underrate the importance of dealing with these very low probability but extremely high impact possibilities. I mean, if you sort of think carefully about the value of our big cities or the value of our civilization, we are probably under investing both in detection and the development of technologies to deflect. So that's just an observation about how we tell the story.

01:29:01.360 - 01:29:56.360

My question has to do with the list of possible approaches to deflection, which listed only three, kinetic impact, gravity tractor, and nuclear. And I'm just wondering, I have not actually gone back and read the Academy report, but it strikes me that there is, at least in principle, another possibility, which is to use, instead of a gravity tractor, use a thrust push. In other words, bring to the asteroid a thruster of some kind that could exert a thrust over a period of time. Obviously, given the cost of transporting stuff into space, probably a chemical rocket would not be very attractive as a thruster.

01:29:56.360 - 01:30:19.360

I can imagine a solar ion thruster, for example, might be interesting. Has that been ruled out on the basis of detailed calculations that indicate it just is not as attractive as, say, a gravity tractor? Yeah. With the NRC report, those three techniques are determined to be the most viable at this time, given our technology.

01:30:19.360 - 01:30:59.360

The difficulty with a thruster, the complexity of trying to thrust, is that these objects, particularly the smallest ones, are spinning very quite rapidly. And so you've got to have, if you're going to, you know, land on the object and try to thrust against it, you've got to have some way of compensating for that. Either, you know, a pulching of the thrust at the right moment, which decreases the effectiveness of it considerably. Or you have to despin this tens of ton mass, which may sound simple, but when you really look at the numbers, it's kyclinplex.

01:30:59.360 - 01:31:21.360

So I wouldn't rule that out, but it is considered a more complex approach than the three that we talked about. Okay. Well, let me close this session. Yeah, sure. The OSTP-led working group that has been set up to address this challenge, Damien, has five components, you know, detection, deflection, disruption, mitigation, recovery, and response.

01:31:21.360 - 01:31:49.360

And I think when we, as part of developing the strategy and the action plan for near-earth objects, one of the gap areas is likely the issue that you have, which is, you know, are there other solutions that need to be looked at, that need to be developed? And, you know, there's many suggestions in the community. I mean, one elegant one is painting the asteroid and using the thermal gradients to move it. So all of those are the ones that may get looked at as well.

01:31:49.360 - 01:32:17.360

Great. Well, again, thanks to the whole panel for a very informative set of presentations and a good Q&A. We'll now break for 15 minutes. And while we are a little bit behind schedule, that's not such a serious problem because I understand we have no public comments or one public comment and abundant time set aside for that. So let us return in 15 minutes.

01:32:17.360 - 01:32:39.360

Thank you very much. Thank you for your continued interest. Bill Press is going to moderate this next session, starting with the introduction of the panel. Bill, as most people in the room know, is one of our two PCAST vice chairs and is much more knowledgeable than I am, certainly.

01:32:39.360 - 01:33:02.360

I'll let the others speak for themselves on this particular topic. Bill. Thank you, John. Well, after coming close to destroying the world in the previous session, it's nice that I don't know what could be more upbeat than the idea that we can make money out of pure bits.

01:33:02.360 - 01:33:40.360

Just money from nowhere. This session is on cryptocurrencies, which in the popular imagination usually goes by the name Bitcoin. But I think as we're going to see, there's a much more general and fundamental idea here, something that has the possibility of revolutionizing not only things relating to money and finance, but a broader set of functions in our civilization as long as it's not destroyed by the asteroid. We have four very distinguished speakers, and let me just introduce them by their titles.

01:33:40.360 - 01:34:22.360

PCAST members, you have their full bios in front of you, and they're going to speak not in the order that they're seated, but they'll speak in this order. Joseph Bonneau is a postdoctoral researcher at Stanford and a technology fellow with the Electronic Frontier Foundation. Jerry Britto is executive director of the Coin Center, a not-for-profit study center. Simon Johnson is Ronald A. Kurtz, professor of entrepreneurship at the MIT Sloan School of Management, and Tim Grant is managing director of R3CEV, which I'll wait for him to speak to tell you what that is.

01:34:22.360 - 01:35:04.360

So why don't we just start right in. Each speaker will speak for maybe seven or eight minutes maximum, and that will allow us a lot of time for conversation that we want to have with you. So, Joe, please take it off. Okay. Yeah, thanks for the intro. It's great to be here. So I guess just quickly by way of background, an academic researcher I've been looking at Bitcoin for about the past five years. From the point in time when to find out anything about Bitcoin, you really had to start digging into IRC channels and mailing list archives and weird places on the internet, look at source code.

01:35:04.360 - 01:35:41.360

Now I can say on the academic side, excitingly, we've developed course materials. We have a Coursera course that I worked on with Ed Felton and some other colleagues at Princeton. We have a textbook coming out. There's a whole literature, a whole academic literature, research papers on Bitcoin. So in the last five years, it's amazing how it's gone from such an obscure field of knowledge to we have some good technical resources out there. Lots and lots of people. I taught a course at Stanford this year and we had 150 undergrad students show up to learn about cryptocurrency, so it's a really, really exciting topic within computer science.

01:35:41.360 - 01:36:16.360

So that's the good news, I think, is that we're gaining much more fundamental understanding. This is really sort of a technology that came from way outside of academia. Obviously, you know, on a personal level, the story of how Bitcoin got created is fascinating by this anonymous technologist. We don't really know who made it, et cetera, but it came with really no proofs, no definitions. It wasn't sort of a very carefully thought through computer science system at the time it was launched, so we've had to develop a theory after the fact.

01:36:16.360 - 01:36:56.360

So that's the good news is that we've made a lot of progress, our technical understanding of what cryptocurrencies are, what blockchains are has expanded a lot. And then I'll say kind of the other side of the coin there. We still don't really have good definitions of these things and the terms get thrown around a lot in the popular discourse and people use them to mean very different things. So I'll just try and give a quick technical definition of some terminology because I'm sure that we'll hear the terms probably cryptocurrency and blockchain a lot today. When people talk about the whole space, those are usually the two overview terms that they use.

01:36:56.360 - 01:37:24.360

And we don't have a universally agreed upon definition of what either of those things are, so they're both sort of closer on the spectrum between buzzword and technical definition. They're still closer to buzzword. We'd like to wrangle them to something we have a technical definition for. So the definition I would give for cryptocurrency is it's a digital currency. Digital currency just means that it's represented in bits. There isn't a paper artifact or anything else that represents who owns the value.

01:37:24.360 - 01:37:51.360

And a cryptocurrency changes hands through the use of some cryptographic algorithm. Most of the time that means that you do a digital signature to say that you want to send value from one person to another. There are other ways to do it cryptographically, but usually we mean it's some digital artifact that has value that you compute a digital signature to send from one person to another. So note that there's a lot of other digital currencies that aren't cryptocurrencies.

01:37:51.360 - 01:38:20.360

There's things like everything from gold in the World of Warcraft game to you could argue that the way we use credit cards most of the time, they're essentially digital currencies because we're not interacting with the physical artifact as much. So that's really the power there is that we can use cryptography directly to transfer money from one party to another and you do it in a permissionless way. If you can compute the cryptographic algorithm, then you're able to send the money. Which sounds like sort of a very small step who cares about that fact.

01:38:20.360 - 01:38:59.360

But when you, you know, we've kind of realized slowly that that's really, really powerful because that means that as the result of some cryptographic protocol, you can automatically transfer money. And it gives us this whole new toolbox when we're designing online protocols for everything from, you know, designing an auction online to two people playing chess online. You can have the results of that interaction actually move value between the two parties and you can do it automatically without having to rely on some third party. So that's, you know, cryptocurrency to me what's really powerful again is that we're using cryptography directly to, you know, to transfer money between parties.

01:38:59.360 - 01:39:22.360

The other thing that we'll probably talk about I'm assuming is blockchains. So on one level, a blockchain is really just a data structure where we have, you know, an append only log. So an ever growing log of data, we can keep adding to it. A cryptographic hash function is used so that, you know, we constantly have a snapshot that captures the entire history of the system.

01:39:22.360 - 01:39:49.360

But I think it's a little bit funny to call that a blockchain because the, that idea dates back at least to the early 1990s. It wasn't called a blockchain when it was first proposed, but sort of the earliest cryptography papers that proposed this are 25 years old now. And they weren't proposed in the context of currency. They were just proposed in the context of, you know, maintaining a ledger of data for historical archiving purposes like that.

01:39:49.360 - 01:40:24.360

So I think what, you know, what goes from the pure data structure to something that we would call a blockchain is that there are precise rules about what kind of data you can add. So the technical term that you would sometimes use would be a replicated state machine, which means that you have some state of the system. And there are specific rules for how that state can change to another state that everybody can verify. So it's replicated in that everybody is running an algorithm that checks for valid state transitions and can check that the blockchain is only having data added to it according to the rules of some state machine.

01:40:24.360 - 01:40:50.360

And in Bitcoin, those rules are fairly simple. They're just basic sanity properties that you'd want in a currency that, you know, you can't create or destroy value and that transactions have to actually be signed. People have proposed much more powerful versions of this blockchain. Since then, you know, it started with a few years of people saying it would be great if we added a little bit of functionality to the state machines that we could represent more complicated things.

01:40:50.360 - 01:41:19.360

You know, famously, people came up with name coin because they said what if we could do a distributed naming system on top of this blockchain. And now we've gotten to the point with Ethereum and other proposals where the state machine is essentially programmable so people can take any program. Technically, we say that the system is turning complete. So any program that computer scientists know how to write down, you can represent in the rules of the state machine and everybody can check that the states are being represented.

01:41:20.360 - 01:41:48.360

So that represents, you know, in some sense, the end game of power. We now have this blockchain that can represent sort of arbitrarily complicated transactions and protocols between different parties. Although we're still really just scratching the surface of how to program that, how to do useful things with that. So on the academic side, that's where a lot of the excitement is now is figuring out more complicated protocols besides just payment that we can build using a more powerful blockchain like Ethereum.

01:41:49.360 - 01:42:07.360

So I just wanted to kick off by saying, you know, just those notes on terminology. I'm sure a lot more questions will come up. But, you know, at a technical level, that's what I see as the big steps forward that we've made and are thinking about this space. And I guess I'll turn it over to Jerry with that.

01:42:09.360 - 01:42:25.360

So as Bill mentioned, my name is Jerry Brito. I'm the executive director of CoinCenter. CoinCenter is an independent nonprofit based here in Washington, D.C. And we're focused on the public policy issues facing cryptocurrencies like Bitcoin and distributed decentralized competing platforms like Ethereum.

01:42:25.360 - 01:42:46.360

And so given that vantage point, what I'm going to do today is sort of give you a high level overview of what are the regulatory questions? What's the regulatory landscape surrounding these technologies? And so the first question we have to ask ourselves is, well, why are governments interested in cryptocurrencies? We've seen a lot of activity where policymakers and regulators are interested in these.

01:42:46.360 - 01:43:13.360

That's for a couple of reasons. The first is that the currency, the token that Joseph was describing, you know, for the first time, these technologies allow us to have scarcity online, provable scarcity. These tokens can be limited in a certain way where you can show provably that when I transfer one to you in a peer to peer fashion, that is a token that you're getting. And this means that the token can represent anything.

01:43:13.360 - 01:43:27.360

To date, these tokens have tended to represent money, right? So one Bitcoin today is worth about $440. That's sort of an obvious application of a token that can be transferred in a peer to peer fashion. And so as because it's so obvious, it's sort of the first application.

01:43:27.360 - 01:43:48.360

These tokens can represent anything. It can represent a particular house, a particular car, an ounce of gold, a share of stock. And again, they can be traded in a peer to peer fashion. And so as a result, financial applications that are built using cryptocurrency networks potentially implicate consumer protection, tax, securities, and other regulations.

01:43:48.360 - 01:44:11.360

And so these are all have drawn the attention of policymakers. Cryptocurrencies, as Joseph said, are also open permissionless networks, a lot like the internet, and they're also decentralized, which means that they lack intermediaries. And so thus they're censorship resistant. Now permissionlessness and censorship resistance.

01:44:11.360 - 01:44:52.360

And by the way, when I say permissionlessness, we've said those words before. What I mean is this, is that previous to the invention of Bitcoin, the first cryptocurrency, if somebody wanted to design or develop a financial application or an application that used a scarce token, they would have to get the permission of a third party intermediary like PayPal or Bank of America, typically through an API, to build that application. Today, because these networks are open and permissionless, they can develop simply by building the application and launching it on a network, much like somebody could do, Mark Zuckerberg could launch Facebook from his dorm room.

01:44:52.360 - 01:45:21.360

And so permissionlessness and censorship resistance are exactly those ingredients that allow for innovation to flourish through experimentation, and they allow also for innovative peer-to-peer uses like microtransactions that weren't really possible before. These same features, however, have attracted illicit users. And so cryptocurrencies have become the payment rails for online drug markets and ransomware. And as a result, there is a lot of scrutiny from law enforcement agencies.

01:45:21.360 - 01:45:55.360

Those are sort of the two main areas where we see interest from government. And the important thing to notice here is that we saw similar policy challenges emerge along with the original permissionless network, the internet. There were questions about consumer protection, about tax, about intellectual property, and there were issues related to illicit uses, etc. But as the framework for global electronic commerce, which laid out the Clinton administration's policy for the internet in July of 1997, it said governments should recognize the unique qualities of the internet.

01:45:55.360 - 01:46:24.360

And it went on to say that the genius and explosive success of the internet can be attributed in part to its decentralized nature and to its tradition of bottom-up governance. Accordingly, the regulatory frameworks established over the past 60 years for telecommunication, radio, and television may not fit the internet. Existing laws and regulations that may hinder electronic commerce should be reviewed and revised or eliminated to reflect the needs of the new electronic age. And we're very lucky that that very forward-thinking policy was put into place.

01:46:24.360 - 01:46:55.360

And today, this same approach is advisable for cryptocurrencies. And luckily, government to date, certainly the U.S. government, has grasped the potential of this technology. Really, big attention to cryptocurrencies began in the fall of 2013 with the first congressional hearings on cryptocurrency. These were two sort of paired senate hearings.

01:46:55.360 - 01:47:31.360

And industry, academia, government all participated in these, and the outcome was very positive. There was an understanding that this technology really could revolutionize finance and not just finance, anything that uses a ledger, which is anything that uses a database, which is just about anything. And the great thing about these, the outcome of these hearings is that there you had the regulators on point at the moment, especially the Treasury Department, announced that there are many, and this is quote, there are many legitimate uses of these technologies, and the virtual currencies are absolutely legal.

01:47:31.360 - 01:47:47.360

And they're very innovative and important for the economy. So that was sort of an overhang. So when Bitcoin began in 2009, there was always a question of, well, you're creating money out of thin air, is this legal? That was sort of put to rest here with these hearings in 2013.

01:47:47.360 - 01:48:12.360

But we remained or all of a lot of different policy details. And there are two major areas of regulatory activity that we've seen today that has to do with consumer protection, on the one hand, and on the other hand, financial surveillance. So I'll talk first about consumer protection. So in 2014, you had the implosion of the largest Bitcoin exchange that was called Mt. Gox.

01:48:12.360 - 01:48:42.360

And as a result, you had the loss of about $600 million in assets by consumers. And this was the function of, because Bitcoin was so new, it was something that really was sort of being developed by enthusiasts, by computer scientists who were having to go through those forums and crazy Reddit posts to find out how to build it. There's really no good way of exchanging it, of buying and selling it or sending it to each other.

01:48:42.360 - 01:49:01.360

And this one company called Mt. Gox sort of emerged as the one exchange, and it gained sort of a network effect. But this exchange was seemingly built by amateurs. And as a result, it collapsed. And this brought a lot of attention of, well, what are the regulatory standards here?

01:49:01.360 - 01:49:30.360

What we've seen since then is that today you have at least a half a dozen, if not more, serious regulated Bitcoin exchanges that are licensed and that are backed by serious investors and have professional management. And part of this is because of the licensing regime that has been put in place. Typically, if you're not a bank, but you're taking custody of consumer funds, you're going to be subject to state licensing, state money transmission licensing.

01:49:30.360 - 01:49:51.360

And today you have about a dozen states that are actively either developing new legislation to create new licenses for digital currencies, or they're amending their existing money transmission licenses in order to accommodate these. So New York sort of was famously the first state to create an entirely new license called the BIT license, where companies are taking custody of consumer funds.

01:49:51.360 - 01:50:23.360

But you have other states like Texas and Kansas that have applied their existing money transmission licensing to these companies that operate in this space. And what's important to understand about the licensing here and the reason that we have it and how we apply it is that the risk is presented when you have a company that takes custody of consumer funds, whether that's dollars or whether that is Bitcoin. And if you're holding onto somebody's Bitcoin, essentially holding their cryptographic keys on their behalf, you're being put in a position of trust.

01:50:23.360 - 01:50:51.360

And at that point, you need to be licensed. And licensing requires background checks, requires that you post a bond, requires that you get permission before you develop new products, et cetera. Now, there's some issues with this, which is the following that we need to be cognizant about. Financial services that could previously only be offered through a full custodial service can now be offered without taking custody of consumer funds,

01:50:51.360 - 01:51:06.360

and thereby posing little or no risk to consumers. Let me give you two simple examples. One would be wallet software. If I wanted to hold Bitcoin, one way I could do it is to have a relationship with a Bitcoin wallet company where I give them my Bitcoin,

01:51:06.360 - 01:51:25.360

they hold it for me and I ask them to transact on my behalf the way I would do with PayPal. So it's a very traditional model, certainly licensed. But there's also the possibility where I can download software created by a company. They simply create software, I can download the software, and I can have full custody of the Bitcoin myself on my computer.

01:51:25.360 - 01:51:51.360

Even if that software is being delivered through a browser, and even if you, when looking at it, it looks very similar to a custodial service. Those kinds of services really increase consumer privacy, increase consumer protection, because there's no company that can run away with money. And so as a result, they do not add risk, do not create risk, they reduce risk. And so we should make sure that licensing is appropriately avoided there.

01:51:51.360 - 01:52:17.360

Money transmission licensing can also present a barrier to innovation, because at number one, you need to get 50 licenses before you can operate. You have to get a license in each and every state in which you operate, and to date, these licenses are not forthcoming, and they've been inconsistent, and it also potentially cancels out the advantages of having an open permissionless platform. This is in contrast, for example, with other countries.

01:52:17.360 - 01:52:38.360

The EU, for example, has easy passporting of licenses. When you get a license in the UK, you're good to go in every other state in the European Union. So that essentially is consumer protection. The way that we deal with consumer protection issues is that we license, actors are taking custody of funds,

01:52:38.360 - 01:53:02.360

and hopefully we exclude from any licensing requirements those companies that are building out solutions that do not take custody of funds. The next big area is financial surveillance. And one important thing I need to note off the bat here is that a cryptocurrency like Bitcoin is not anonymous, despite what many people read in the newspapers.

01:53:02.360 - 01:53:28.360

It is better to call it pseudonymous. In fact, the very idea of Bitcoin's blockchain is that there's a record kept of every single transaction that is carried out over the network. So to give you an example of a truly anonymous transaction would be a cash transaction. If I'm selling a bicycle at the flea market and Joseph comes along and we agree on a price of $100,

01:53:28.360 - 01:53:42.360

he gives me $100 bill, I give him the bike, we part our separate ways. I don't know his name, he has no my name. There is no record kept of that transaction to date the amount of time, the purpose, none of that is recorded. That is a completely anonymous transaction.

01:53:42.360 - 01:54:01.360

On the other hand, you have something like, say, a credit card transaction, which is a perfectly identified transaction. If Joe were to pay me with a credit card, his bank knows his name information, my bank knows my name information, and there's a record kept of the date, time, and amount of that transaction. On something like the Bitcoin blockchain, it's a little bit of both.

01:54:01.360 - 01:54:22.360

It's a pseudonymous. There is absolutely a record kept, the date, time, and amount of the transaction that's sort of necessary for the system to work. But our names are not necessarily tied to those transactions, certainly not on the public blockchain. And so this presents sort of a dilemma for law enforcement who are used to being able to have somebody's name that they're investigating,

01:54:22.360 - 01:54:42.360

being able to go to their financial institution and say, can you show me all of Joe's financial records, show me everybody he's transacted with, with a subpoena or a search warrant. That's not possible on the Bitcoin network in the sense that you don't know, just looking at Joe's name, you don't know what transactions are there.

01:54:42.360 - 01:55:07.360

It's a bit reversed. With Bitcoin, you can see a transaction has happened, but you don't know who is attached to this transaction. You can then go to a financial intermediary, something like an exchange, and say, can you reveal to me who is the name, who is the person associated with this Bitcoin transaction, and they can do that.

01:55:07.360 - 01:55:29.360

So it's sort of the inverse of what law enforcement has traditionally been used to, but still there are tools there for them to investigate. And so in order for them to have access to information, they need to, intermediaries like exchanges, need to be required to record information about their customers.

01:55:29.360 - 01:55:52.360

And so in March of 2013, the Financial Crimes Enforcement Network, which is part of the Treasury Department, issued guidance saying that the crypto currency exchanges are subject to the Bank Secrecy Act. And so that means that they must register with Vincent, they must complete know your customer. Could you wind up in maybe a minute or two?

01:55:52.360 - 01:56:06.360

I'm so sorry, I haven't taken a long time. And have basically create records that they can then have make available to law enforcement. So today law enforcement is able to complete investigations. So I'll stop there.

01:56:06.360 - 01:56:17.360

I'll also say the tax questions have also come up, but they've been addressed as well. And I'm happy to answer those questions about those. Thanks. And we can come back in the question period to other things you might want to say.

01:56:17.360 - 01:56:23.360

So Simon, please. Thanks very much, Bill. My name is Simon Johnson. I'm a professor at MIT.

01:56:23.360 - 01:56:38.360

And I used to be the chief economist at the International Monetary Fund. And I think, Bill, the truth of the matter lies in the second part of your joke about why we're here today, which is the extent of the profound change that we're looking at. And I think your timing is impeccable.

01:56:38.360 - 01:56:53.360

Because as Joe emphasized, that the science is still evolving, and yet we can already see some pretty big implications. And let me try and tackle or put forward five. I think the right question always in this sort of discussion is,

01:56:53.360 - 01:57:08.360

what is the problem that we're trying to solve? What Joe and Jerry have told you is there is a new technology, a arguably better still developing technology for organizing transactions in a decentralized manner. Something as fundamental as payments, which we have for a long time,

01:57:08.360 - 01:57:19.360

600 or 800 years, have run in a centralized way. Through a centralized trusted party, we can now run that in a decentralized way. And Bitcoin is the proof of concept. What else could you do with this?

01:57:19.360 - 01:57:31.360

What are people doing? And of course, I think you've all picked up on the fact that very large companies of all kinds, are investing heavily in this technology. There's a lot of hype, and hopefully we can strip away some of that hype with you today.

01:57:31.360 - 01:57:49.360

But I think there's some really important realities changing. Well, the first use case is the most obvious, and the one that I think is always comes up at the top of this kind of agenda, which is situations where you have a central authority, central authority doesn't work very well, it's corrupt, it's not efficient, they screw up in some way,

01:57:49.360 - 01:58:00.360

now we can bypass the central authority. So a leading example would be land registries. We've known for a long time in many countries land registries don't work well, you can pay someone to change the land record when political power changes,

01:58:00.360 - 01:58:12.360

first thing they do is change who owns assets. Can we now organize land records in a more decentralized fashion? The answer is absolutely yes. Now, on the point about, let's not exaggerate, let's not get carried away,

01:58:12.360 - 01:58:26.360

there are a lot of other problems with the creation of functioning land registries that are not solved by a blockchain or by decentralized technology, but there's no question, look around the world, look at what people are doing, we have at least a catalyst, we have a change of the conversation,

01:58:26.360 - 01:58:42.360

we have a new focus in these discussions, and we'll hear about some of this when we talk about financial markets with our three in a moment. The second use, which is getting much less attention, but I think it's extremely timely for you to think about it,

01:58:42.360 - 01:58:56.360

is situations where we don't have a central authority and we don't want a central authority or isn't going to be one, but we have dispersed records and we want to integrate those records and we want to manage those records better. We're extremely bad at this and you can look at the US healthcare system as one example

01:58:56.360 - 01:59:08.360

and you can look at the VA system for an even more painful example that matters a lot to the federal government right now. So there's a technology being developed at MIT called MedRec, which I highly recommend to your attention,

01:59:08.360 - 01:59:22.360

which is a decentralized way using a blockchain type approach to manage access to medical records while the primary records still remain where they have to remain according to law, but you can access them, you can manage them in a fundamentally different way,

01:59:22.360 - 01:59:39.360

and you can manage this network in a way that's absolutely inspired by this technology and uses Ethereum actually as a specific technology. The third use case is one where we see big companies, and I should have said we have a fairly substantial project at MIT

01:59:39.360 - 01:59:54.360

working with private sector people on these use cases and trying to think about not who should be the right vendor or even what technology you should pick, but where's the social impact and what should we care about from a policy perspective.

01:59:54.360 - 02:00:07.360

Supply chain management, how you move goods across borders is a very complicated and messy problem. So in these situations there is a trusted set of interactions already, the extent of trust may vary.

02:00:08.360 - 02:00:23.360

There is not a lot of automation in this kind of transaction, we move a huge amount of value, we have to coordinate between private sector firms, we have to coordinate with customs, we have to coordinate with freight forwarding,

02:00:23.360 - 02:00:36.360

we have to have trade finance in this, and it's the extent that you can run a better functioning, more decentralized blockchain, perhaps, and we haven't talked about this enough, there's a permission to blockchain.

02:00:36.360 - 02:00:51.360

To be frank, there's a big debate in the community about the extent to which we can or should or will rely on a relatively permissionless Bitcoin type structure, or the extent to which you can build a relatively controlled system

02:00:51.360 - 02:01:13.360

with a defended perimeter and within which there is decentralized interactions. The fourth use case is real-time information. So just picking up the Wall Street Journal today, you can see a couple of spectacular examples of instances,

02:01:13.360 - 02:01:28.360

these happen to be in the financial sector, where executives have taken positions that were not in compliance with the firm's policies, and the risk committees or the responsible management didn't know what people were doing. So what you have with blockchain technology defined appropriately

02:01:28.360 - 02:01:43.360

is a way for people to disclose what they're doing and to provide information on an appropriate basis while protecting their privacy. So you can determine who has access to appropriate knowledge about your investment positions. If you're following, for example, what's happened with Lending Club

02:01:43.360 - 02:01:57.360

or what's happening at Deutsche Bank now, this is highly relevant. And more broadly, for the financial community, there's also a lot of interest and effort going into developing the ability to audit activities and understand what your own firm is doing.

02:01:57.360 - 02:02:14.360

For example, in terms of risk positions of a financial firm in real-time, because again, the structure of information changes fundamentally. It doesn't have to flow up through a hierarchy. It can be available instantaneously through access to the appropriate cryptographic keys

02:02:14.360 - 02:02:33.360

so you know what positions your firm has in all its transactions. The sixth and final use case I would mention is that wider use of this technology, including for payments, is going to allow us to change the structure of contracts in the economy.

02:02:33.360 - 02:02:48.360

So Puerto Rico has a large debt problem. Puerto Rico offered to swap some of its debt for a GDP index bond in the beginning of February. The notional value they said was $25 billion. The creditors said the value of that to them was roughly zero.

02:02:48.360 - 02:03:03.360

Why? Because GDP is a made-up number by all statistical authorities. In fact, there's another committee meeting, I think, down the hall on this today. The creditors have experienced in Argentina, in Greece, in Ukraine that these numbers will be changed when it's in the interest of the issuer to change them.

02:03:03.360 - 02:03:17.360

Well, what we're discussing today is a different way to make payments, a different way to verify transactions in a decentralized way. So it's not a central authority that's telling you what happened, either the government reporting or a firm reporting.

02:03:17.360 - 02:03:32.360

But that data is now much more widely available, and contingent contracts, equity-type payoffs, will become much more broadly available. The last point I would make is that central banks, which are a very important gatekeeper, including in the regulatory situation that Gerry was describing,

02:03:32.360 - 02:03:49.360

are, well, I wouldn't say enthusiastic, but they are positive about this development. They are encouraging innovation in this area because they see potential for moving away from some of our current fragilities. We have a more decentralized financial structure, potentially,

02:03:49.360 - 02:04:04.360

if we have an open enough architecture. We can have a change in the form of our contracts, so away from debt towards something that's more resilient in the face of shocks. And we may be able to reduce transaction costs in some of our more fundamental financial transactions,

02:04:04.360 - 02:04:17.360

which obviously is a very pressing need in many countries. Thank you very much. Thank you. Thank you very much indeed.

02:04:17.360 - 02:04:28.360

And on behalf of the R3 team and our members around the world, we appreciate the opportunity to speak in such a steam company. Look, thanks to my fellow panelists. You've all said a lot of what I was going to say,

02:04:28.360 - 02:04:45.360

which is actually very helpful, lack of definitions, regulatory questions, financial surveillance, the investment in the space, the hype cycle. Let's not, you know, lose sight of the fact there's a huge amount of hype around this. Supply chain, smart contracts, central banks, this is all the stuff we want to talk about.

02:04:45.360 - 02:04:55.360

In the next few minutes, I'm hoping that I can control this. There we go. I'm quick going to spend two minutes on each one of these. What is R3?

02:04:55.360 - 02:05:04.360

That'll be quick. What is a distributed ledger to technology and how does that differ from blockchain? There is a difference. What is the promise of DLT for financial services?

02:05:04.360 - 02:05:12.360

And how are we going to get there? That's really the ultimate question at the end of the day. If any of this is actually going to fulfill the promise, how do we get there and make this into production?

02:05:12.360 - 02:05:22.360

So what is R3? We don't have exploding asteroids. We don't have, I don't have the ability to control this apparently. There we go.

02:05:22.360 - 02:05:38.360

But I do have the world's biggest consortium of banks working on distributed ledger technology. We have 50 member institutions now from all around the world. We have literally some of the very biggest, you recognize the names. The key is that we've come together to try and solve problems.

02:05:38.360 - 02:05:50.360

We're exclusively developing next generation financial transaction network and commercial applications. That's the whole point. And we're doing that on distributed ledger technology, which has a very distinct relationship with cryptocurrencies.

02:05:50.360 - 02:06:00.360

And hopefully that'll become clear by the end of the next few minutes. So what are we doing from here? We were 50 members as of today. We're growing very fast.

02:06:00.360 - 02:06:22.360

Important to note that we're looking to grow to the entire ecosystem of players around the world. So yes, other financial institutions, for profit, asset managers, exchanges, market structure utilities, the likes of Microsoft and IBM and Intel, who we have good relationships with. Small blockchain companies, we have great relationships with all of the incumbent startups.

02:06:22.360 - 02:06:36.360

But perhaps most notable, as was mentioned before, regulators and central banks, we fundamentally believe as a membership that the only way to get this stuff adopted, the only way to get this stuff in production, the only way to realize the promise, is to work with the regulators and the central banks.

02:06:36.360 - 02:06:48.360

And in fact, we're already doing so in a number of jurisdictions around the world. So that's what our three is, in short. What is distributed ledger technology? Now, I'm not going to spend a lot of time on this.

02:06:48.360 - 02:07:07.360

There's two kind of key elements of this paradigm that have been mentioned already. We're moving from a centralized world to a decentralized world that has some certain implications in terms of permission versus permission list that I won't label the point on. In the old world, Facebook's a great example of a centralized network.

02:07:07.360 - 02:07:19.360

All the data is in fact centralized. Nothing resides on the individual computers of the users. On the other end of the spectrum, we have Bitcoin and Ethereum that have been mentioned already, which are indeed fully decentralized.

02:07:19.360 - 02:07:38.360

Without putting a philosophical stake in the ground, this is the world that we're all debating where we end up and how we realize the potential benefits of this technology. So swiftly moving on, let's just call out a couple of key concepts

02:07:38.360 - 02:07:50.360

that hopefully will resonate and you can walk away with these ideas. Cryptography is key. It's been mentioned before. It's a huge part of everything that we're doing. Non-repudiation or immutability is key.

02:07:50.360 - 02:08:08.360

The ability to be sure that the information is real, let's just call that truth as well. Smart contracts have been mentioned. This is going to be potentially a significant benefit to the entire financial system and beyond to be able to have automated execution of business logic at its core.

02:08:08.360 - 02:08:21.360

Shared ledgers or distributed ledgers, the idea that the right people see the right information at the right time and that it's not decentralized, that's a huge part of this. Consensus, the idea that we can settle upon the truth

02:08:21.360 - 02:08:37.360

and that the right people can be clear what the truth is. What's really key and it's been mentioned already is that none of this is new. On its own, none of these ideas are particularly innovative. It's the way in which they've come together that is particularly impressive.

02:08:37.360 - 02:08:53.360

If you haven't read Satoshi Nakamoto's paper from 2009, it's a genuinely beautifully written research piece and we urge all of you to read it. That was really a bit like Henry Ford pulling together the internal combustion engine and mass manufacture.

02:08:53.360 - 02:09:06.360

He didn't invent either of them but he put them both together to great effect and that's an interesting little analogy. This following slide gives you a sense of the importance of the steps that we're potentially moving into.

02:09:06.360 - 02:09:25.360

Before the internet, the notion of processing information at its core was done in a way that where we weren't sure that everyone was doing it in the same way. Let's take a really simple example which is one company invoicing another company. Company A has a ledger and that has information about who's owed what.

02:09:25.360 - 02:09:39.360

They produce a document called an invoice. They send that invoice through some communication mechanism and the reverse happens at the other end. Before the internet, that communication part was really potentially fraught with inconsistencies.

02:09:39.360 - 02:09:50.360

What the internet did was remove that processing component and we've all seen the extraordinary results. The next step is the truth step. It's the step around perception.

02:09:50.360 - 02:10:05.360

That fact that got sent over that may have been put into the ledger in the second company may have been put in the wrong way. Therefore, we're not sure that what's on one ledger is exactly the same as on the other and that's when we have your reconciliation issues

02:10:05.360 - 02:10:21.360

and that's what this step is looking potentially to solve. It's to allow us to reach some form of truth. It sounds perhaps a little bit pie in the sky but that's the gravity of what we're potentially dealing with here.

02:10:21.360 - 02:10:42.360

One unique approach and I should be very, very clear. R3 is building technology but we don't believe there is only one ledger or blockchain that will ever exist. There will be many, they will work together, we'll need to inter-operate. This is just an example of the kind of problems that we're trying to solve in the financial sector.

02:10:42.360 - 02:11:02.360

Our approach is called CORDA and the key parts are we don't think you need to share all of the data all of the time. Certain other paradigms do have all of the nodes, have all of the data all of the time. Financial markets and financial institutions don't believe that to be necessary so that's a key design decision that we've taken.

02:11:02.360 - 02:11:16.360

We've designed this with regulators in mind. We are fundamentally a regulated industry globally. We don't believe that we're going to get away from that. We don't believe that that's a bad thing except when it comes to capital that we might have to hold.

02:11:16.360 - 02:11:36.360

This is designed with regulators in mind such that regulators and anybody else who needs access to information will always have access to that information. Legal pros, we're already getting to the point in this march towards reality that we need to be sure that smart contracts actually represent something real.

02:11:36.360 - 02:11:50.360

In some respects smart contracts is a bit of a misnomer. It should be really smart logic. It's a smart contract when there's legal pros in there and it has some contractual basis in reality. I'll move on from this page.

02:11:50.360 - 02:12:07.360

The idea was to try and give you a sense of the kind of problems that we're definitively trying to solve as a group. What's the promise of this technology? What is it going to do? If you ask around you'll get pixie dusk and unicorns will be waved

02:12:07.360 - 02:12:18.360

and within a couple of years we'll be saving hundreds of billions of dollars and all kinds of new business models will sprout up. The truth is it's going to take a lot longer. It's going to take a lot of collaboration but if we do it right,

02:12:18.360 - 02:12:37.360

as this slide is just one small slice of the number of use cases that could be applied, Simon went into some of these. I'm not going to go into all of them but the point is for our financial institutions, for regulators, for individuals and beyond finance we're aware we're focused on finance

02:12:37.360 - 02:12:55.360

but beyond this technology has potentially significant implications for reducing costs and increasing business opportunity. Another way of saying it is we're going to reduce operational risks and increase efficiency. Reduction of manual and error-prone processes.

02:12:55.360 - 02:13:05.360

Think of that example that I mentioned earlier on. The ease of validation of financial transactions through smart contracts, that's already been mentioned. I think that's intuitively very clear.

02:13:05.360 - 02:13:20.360

If we can make it work it could be very useful for all of us. The real-time legal entity, auditor, regulator access, the market surveillance part that Jerry talked about, very much front and center. And knowing your client, knowing identity, knowing who you are,

02:13:20.360 - 02:13:35.360

this has actually very real implications beyond finance in terms of identifying individuals and companies and other entities. So the benefits are potentially huge. I'll finish very briefly if I can get to the next slide.

02:13:35.360 - 02:13:48.360

I'm going to skip a couple so I realize I'm running out of time. Where do we go from here? There's a couple of slides there on the Lehman crisis which definitively could have been avoided had we had this paradigm.

02:13:48.360 - 02:14:04.360

These are the big design challenges that are left for us to solve. Let's not pretend we've solved them as an industry or as a technological community. We need to know how everything's going to inter-operate. We need to know how to scale this stuff so that it operates at the scale that we need in an enterprise level.

02:14:04.360 - 02:14:16.360

It has to be secure. It has to be private. It has to be identity and systems integration. The point is none of this has been solved yet. We're working very hard and there are a lot of people putting a lot of time into it.

02:14:16.360 - 02:14:32.360

So the final point and the final slide is simply to say that the only way to do this is through collaboration. If I can get to that final slide. There you go. Our fundamental philosophy is to bring everyone to the table and work together.

02:14:32.360 - 02:14:48.360

We're working on a lot of live projects with everybody you see on this slide. We have a lab and research center that we want to work with academics and governments as well and ask the questions and solve them and have a roadmap to the future. Thank you very much.

02:14:48.360 - 02:15:07.360

Thank you. I see Craig's flag first and then Jim's and then my Craig and Eric. Thanks. That was great. Joe could answer this question or any of you but I haven't looked at this.

02:15:07.360 - 02:15:29.360

A lot of this obviously depends on cryptography. What's the particular nature of the cryptography that's employed now only for the identity part and for the core of the ledger capability? And what do you see downstream as the issues associated with any vulnerabilities associated with that?

02:15:29.360 - 02:15:44.360

Particularly for example, is it quantum hard? Are people looking at that? Yeah, sure. I can answer that. One thing that's kind of interesting about Bitcoin actually is that there's no encryption.

02:15:44.360 - 02:16:07.360

You get to sort of skip that whole aspect of crypto. There's really two specific algorithms in Bitcoin that are the workhorse of the entire thing. The hash function, SHA256, which is standard, the most widely used is the NIST standard. And then there's ECDSA, elliptic curve signatures, which are also based on a NIST standard.

02:16:07.360 - 02:16:31.360

A lot of the time people say it uses the same ECDSA that's used in TLS for secure web browsing all the time. It's actually a very slightly different version that's used in Bitcoin but very closely related technically. So it uses very standard crypto. The crypto algorithms get broken for sure.

02:16:31.360 - 02:17:01.360

If you asked me to vet my life that there would be no successful crypto analysis in the next 25 years, I wouldn't do it. Quantum computers are one way that that might happen, but there's also classical ways that crypto systems get broken. The good news is that Bitcoin and most of these other systems are probably resilient and that they're able to change on a several year time frame. And it would be a little bit of a leadership challenge that I'm not sure the community is totally up to right now.

02:17:01.360 - 02:17:16.360

But if you told me that these systems were these underlying cryptographic primitives would be broken two years from now, we could roll over and switch in that time period. So certainly like the quantum. Assuming you knew one to switch to. What?

02:17:16.360 - 02:17:37.360

Assuming you had one to switch to. Yeah, exactly. But I mean, you know, the quantum like post quantum cryptography is a major research challenge for all, you know, a lot of the crypto research community right now. NIST is going to run a competition and five years will have some standard post quantum signature algorithms.

02:17:37.360 - 02:18:09.360

And, you know, in the event that quantum computers start getting closer and it's going to be we're very, very far and there's going to be a lot of milestones that will hit along the way we'll get a sense that they're closer. It's very unlikely that one will just be invented out of scratch that works, you know, six months from now. Yeah, you know, given a few years of time to switch, it would be easy to migrate. So of all the problems in Bitcoin, I think catastrophic cryptographic failure is very, very low on the agenda.

02:18:09.360 - 02:18:34.360

We should mention just just make sure everyone's aware that a lot of the mining of the verification transaction Bitcoin takes place in China, about 40% is one estimate. And there are also concerns that the mining pools are well, they're certainly more concentrated than the original aspiration, which is to run a completely decentralized system. So this is separate, but perhaps not ultimately unrelated to what Joe's. Yeah.

02:18:34.360 - 02:18:58.360

Well, yeah, I mean, I wouldn't say that that's doesn't that would be true no matter what cryptographic algorithms we use. That's just a matter of where electricity and, you know, computer chips are achieved to build and run. There's just one follow button. I mean, there's the real issue in my mind is if I believe everything you said, once you get these things deployed at scale,

02:18:58.360 - 02:19:23.360

you know, how you transition from one generation to the next is very, very tricky. And so while you could say conceptually, it's easy to wave your one and say, OK, we'll do version two and it'll have this other algorithm. How you get all the assets from one to the next, you know, also seems like it could be tricky. Yeah, so that there is a way to do it.

02:19:23.360 - 02:19:58.360

I mean, we would just sort of within any Bitcoin or any other crypto system, we would have a new type of address that you could send money to that would represent a public key in a new signature algorithm. And there would basically just be a time like, you know, if you when when you change away from the fronk to the euro, you have to say, you know, you have two years where you have to exchange for the new thing, after which, you know, in this case, it might be dangerous to keep it around. There's probably going to be some old systems that never move it over and maybe they'll get attacked, but hopefully, you know, the majority of people will will change over and do course.

02:19:58.360 - 02:20:48.360

But it just, yeah, the problem is not so much, you know, formatting and having a way to do it is, you know, there would be some tale of people that just never upgrade and then probably lose their money when the crypto system is broken. Perhaps to start another angle to that. Obviously, we represent, I have no idea how many billions of dollars of IT spend every year across the world in current ingrained legacy systems. It's definitely true that it's more challenging to innovate when you have, you know, decades of technological debt to fight against. And that's why we, in our three, and I think in the community to hold, and there are more players than just our three, we have to think about systems integration and that change over now, not in five years, and we build that into the plan.

02:20:48.360 - 02:21:17.360

Otherwise, we will simply not be able to do it and large institutions won't make that call. It's just too expensive. Even facing the cyber threat, the banking industry hasn't managed to get itself out of that morass. And, you know, a forced function transition due to it, an unexpected transition here, I think would be even more challenging. So I encourage you to think about it. Yeah, I mean, I could add one more quick note, which is to say we've gone through this on the internet many times.

02:21:17.360 - 02:21:56.360

You know, we've tried to transition away first from MD5, then from SHA-1, and those have both been very disappointing. The internet has been very, very slow to move. Browsers have tried to force websites to stop using these so they could turn it off and people have always dragged their feet. But I do think the economics are a lot different in cryptocurrencies because you can basically, it's not a matter of somebody telling you will cut off access. It's a real threat saying like your money will be gone if you don't switch. So hopefully all the different parties that need to switch will do it as opposed to people who don't want to reinstall some new software on their web server because they don't think that the browser manufacturers are actually going to cut off service to them.

02:21:56.360 - 02:22:36.360

But as Simon said, once you move away from currencies to land records and everything else, you know, those incentives all become highly variable and people's ability to process it also highly variable. Okay, I'm going to move us on a very interesting point. Jim Gates is next. Thank you. I want to thank the panel for the briefing and actually this question comes in response to a comment from Tim. You mentioned the Lehman Brothers incident, shall we say. And I'm no financial expert, but most crashes in my study of them seem to follow a general pattern which is they are about extrapolations into the future and that's where you lose value.

02:22:37.360 - 02:23:29.360

From the tool of crash in the 1600s to our crash of our market in 2008, it's always the same pattern about creating financial instruments based on projections in the future which then don't bear the value that they're projected. So how does an occurrence break this cycle? That's a good question. So I appreciate that question a lot actually. So I will speak from some degree of expertise having been the guy at UBS which as you may know was one of the biggest losers of the crash who got the call on Friday the 14th of September from the group CFO and the CRO saying please tell me what our exposure to Lehman Brothers is. So that was a long weekend with very limited sleep. But the reality is that that took me and a large team 48 hours to figure out what that was.

02:23:29.360 - 02:24:29.360

And by the end of that weekend we had nothing more than a basic spreadsheet to tell us what our exposure was in multiple billion dollars to one of the biggest other financial institutions around. So really I think we divorce the market forces and the exuberance and the expected versus the realized value from what happened next. And if we were able to all have a real time completely transparent perspective on what our exposures are to everybody we would have been able to unwind in a much more orderly way. And it would have saved a lot of money actually and potentially could have saved Lehman from going down. A lot of what what what affected Lehman and other big financial institutions was the very fact that they really didn't know quite what their exposures were. And I think I don't think that's pretty controversial eight years after the fact. And I've seen it firsthand. So you know boy would I have liked to have had that button that I could have pressed on September 14 2008 and had all the answers.

02:24:29.360 - 02:25:15.360

But I still don't understand how this impedes the creation of these fraudulent financial instruments that to me. Oh it doesn't. Well I think there probably are elements of of of that. But I'm the way in which you could have really avoided that the Lehman collapse which is the statement I made was you know a lot of other institutions took the same bets and didn't collapse. Obviously in that particular case they made some I think strategic decisions at board level which put them in a slightly worse situation. But had they have had the kind of level of transparency aside how it all got to that point. It may not have gone down. I think that's a very fair statement. I persevere. I rather agree with you. I think this will be big changes for finance and you should follow carefully the lending club scandal which is now unfolding and think about how you could

02:25:15.360 - 02:26:03.360

organize that information differently and how it was managed differently within the firm with supervisors. We could have a better structure. But I agree with you that I was the chief economist in 2007 2008 and none of this is my fault. I should have said that at the beginning. But you're right. The boom bus cycle is not going to go away. We're going to change the nature of it and we're going to change a bit the winners and losers. But that cycles can be with us. Yeah. I could just add there have been numerous examples in the cryptocurrency space already a huge speculative bubbles that have already popped. Lots of different alternative currencies to Bitcoin that were proposed went way up in value. Then people realize there was no real innovation. The value collapsed. So it's happened again and again. People have invented a lot of new ways to have sort of a bubbles within Bitcoin or other cryptocurrency things.

02:26:03.360 - 02:26:58.360

But I think that human nature of people wanting to invest in the next big thing is never really going to change. Good Eric. I was curious about the brief mention of medical applications and how this was going to change medical records. If you could say a little more about that. I think this is. So this is an innovation for MIT. Andy Littman's group at the media lab. And it's still under development. But I think it is absolutely fascinating because it's not going on. We don't have a central authority that's not functioning. We don't just we don't have anything in the middle. And we know what we need to be able to call upon records. Have people with appropriate authorization be able to look at them be able to integrate that information. But we're also not allowed to move data from it was it will reside where it currently resides, which is obviously mandated by law. So I will send you the information that we have on med rec at MIT. And I'll suggest that Andy follows up with you and your team.

02:26:58.360 - 02:27:59.360

I think it's highly relevant for all the things that you work on. For example, please do. Thank you. My question, I guess it comes from something Jerry brought up at the consumer level use of this somehow Jerry was trying to make a distinction as I understood it between whether the value is held by an intermediary or whether the value is held on, you know, my computer here. I don't understand that because my computer here is constantly in contact with the with the web. It's doing negotiations. It can be hacked. I don't really understand any fundamental distinction between hacking my bank and hacking through my bank to my computer. You know, it's just another tentacle of the the octopus and and and I guess another way of saying the question is if I might think I have a wallet on my computer, but I the consumer have no way of doing independent cryptographic checks on whether it's long since already stolen my

02:27:59.360 - 02:28:42.360

Bitcoin and is just stimulating the transactions in a kind of Ponzi scheme, namely for a long time. It does pay my people. But then, you know, when when zero day arrives, everybody loses their money. So so I wonder if you could say something from the sort of naive consumer side. Sure. I mean, I guess I can say in general it's very hard, you know, for not just for consumers, it's very hard for expert computer scientists to be sure that a computing device that they are holding actually does what it's supposed to do. We basically never solve that problem. And there's good reason to believe that we'll never completely solve this problem. It gets into this trusting trust of who made your computer, who compiled the software that's running on it, etc.

02:28:42.360 - 02:29:35.360

And it sort of it sort of never goes down. So when we're doing computer science research, especially security research, we usually assume some trusted computing base that you hope is behaving correctly. And it's probably fair to say that your entire laptop is too big and that we don't really know how to deliver to consumers laptops that they can trust or behaving exactly the way they're supposed to. So there have been some alternative solutions for Bitcoin specifically that try to fix this. So there's a bunch of different companies that sell small special purpose hardware devices whose only job is to keep your Bitcoin keys safe and have some small interface for you to authorize, you know, sending and receiving money. And then the hope is that you buy this specific device and it doesn't you can't install apps that you can't install malware and hopefully that will, you know, be a good steward of your crypto keys that you can trust and put in your pocket.

02:29:35.360 - 02:30:17.360

But in general, yeah, we have, you know, we hope that you can run some software that will maintain your key, but we don't have any great. I mean, that's that's been an unsolved problem in computer science for a long time. Right. And I think what you're getting at is are you just moving the attack surface somewhere else. But here's what you're also doing is that you're creating for the first time the ability for consumers to hold their own funds, which was not possible before. And so this opens up a whole host of new possibilities. I think in the developed world, we lose sight of why this is important because we have very good trusted financial intermediaries that that we can use. And when we deposit our money there, we know they're typically not going to go away.

02:30:17.360 - 02:31:07.360

We know that our currency is well managed and it's not going to be inflated away. And we know that if there is a hack, we're probably going to be made whole because there are laws in other parts of the world. That's not the case. And so there is for consumers a real value to having the ability to hold value for themselves and to accept value directly. So in many parts of the world, just just staying to the narrow payments application, which is just one of many applications of this technology. In many parts of the world, there is no good way to accept electronic payments. In fact, I was recently, it was brought to my attention recently that PayPal does not operate in Puerto Rico. So in Puerto Rico, you could not use PayPal, but Bitcoin is something that all you need to use is an internet connection, a device, and free and open source software.

02:31:07.360 - 02:31:46.360

So you're right that you're always going to be potentially subject to attack, but you have new capabilities that you did not have before as a consumer. So I think that the future of money largely actually depends on what central banks decide to do. And I don't think I think the central banks are going to worry about what you're worried about. And I think that while the future of money is absolutely digital and it's from a consumer point of view, they don't know and they don't care how that digital representation actually operates. I think that the extent to which we actually use Bitcoin literally in our monetary transactions will be relatively small. I think this blockchain technology has massive other profound implications.

02:31:46.360 - 02:32:33.360

And Bitcoin at least brings very healthy competitive pressure onto the banks and onto the central banks to up their game because they're way behind in terms of applying existing opportunity, existing technology, improving how the systems work. But I rather suspect that also linked to the point that Craig made in terms of the vulnerability of our systems, I think the central banks are going to shy away ultimately from the pure cryptocurrency solution. Actually, I totally agree with Simon that I think the impact of cryptocurrencies as daily use consumer payment systems is usually overstated. I think most of the impact is going to be elsewhere. But I guess I wanted to challenge something Jerry said about how the ability to keep currency yourself is something fundamentally new that cryptocurrencies are delivering.

02:32:33.360 - 02:32:55.360

Because I could withdraw all of my personal savings as cash from the bank and put them in my freezer if I wanted to. And most people would tell me that's a bad idea because then I could lose everything if a burglar breaks into my home. Which to me, the analogy is not perfect, but there's a... I should have qualified by saying that in an electronic fashion. Right.

02:32:55.360 - 02:33:31.360

Okay, so the difference is really just you can put your whole net worth either on your laptop or in your freezer or wherever you'd bury your cash. In both cases, there's risks to holding your assets yourself and there's risks to loaning them out to some bank that might lose them. The difference is that in some developing countries, there aren't really great options for putting your funds in a financial institution that's trustworthy. In the West, bank robberies have become a lot less rare, so they're not something that's on most people's radar.

02:33:32.360 - 02:34:03.360

And in the cryptocurrency space, exchanges are much more likely to fail, so the balance of probabilities of keeping your funds yourself versus loaning them out is probably a little different. So I think this is the key point, if I may, because what is cash? Cash is this token issue by central banks that has no value except what people think it's worth. And what's the ultimate full bank in other countries? It's US dollars cash. That's where most hundred dollar bills are held. And so what the central banks are grappling with is they give you this non-digital token, the piece of paper.

02:34:03.360 - 02:34:24.360

They do not give you access to what they could, which is a digital token liability of the central bank. That you can only have through the private commercial banks. The central banks are trying to figure out how to change that and how to give you a robust digital token that could be used globally, and then, you know, you can argue about how that system runs without, I'm not saying that they necessarily want to keep the banks as they are,

02:34:24.360 - 02:34:54.360

but they're very worried that if they do that in a very rapid way, that could be disruptive and we could generate all kinds of issues, including in how the credit system operates. But that appeal, this Joe Jerry discussion here, that's exactly what the central banks, the big ones in the leading countries of the world are grappling with right now. So, Eric Glander, last question. I'm just following this discussion. I'm not sure I've got all of it, but I'm getting very worried about somebody stealing my Bitcoin or the thing in my freezer or wherever it is.

02:34:54.360 - 02:35:31.360

And it occurs to me, is there some way, maybe this is not a well-posed question, but where my currency can be distributed across multiple places, so if it gets stolen from not too many of them, I still have it? Great question. Yes, the answer is yes, there is. Actually, even dating... I'd like to do that. If you could work that out with me after this session. I want the same dollar in multiple places, so that dollar can't get lost.

02:35:31.360 - 02:36:08.360

So, what's interesting, I think this actually is something that is fundamentally new about cryptocurrency. Even dating to Bitcoin, it's built into the Bitcoin design in every system since that coins can be assigned to some quorum of keys. So, you can say here are three cryptographic keys, and if any two of them agree to send the money, then the money is sent. So, unlike with things like cash, which you can't really split, you could put your cash in three different places, and if one of them gets stolen, then you'll lose a third of the cash.

02:36:09.360 - 02:36:39.360

With digital currencies, you can sort of put your keys in three different places, and if any one of those keys gets stolen, you're actually still secure. It doesn't have to be two out of three, it can really be any number out of any other number. It's actually been a research project at Princeton to streamline that technology a little bit and make it more efficient, so I'm glad you asked because we've been working on it. So, that is very exciting, and there's a lot of different models that have been proposed that don't really have good offline analogies.

02:36:39.360 - 02:37:14.360

You can use an intermediary, and you can have them hold a key, and you also hold a hardware key locally. I will say that this poses a lot of regulatory challenges too, because most financial regulation is based around the concept of a token actually being in some physical location. That's no longer necessarily true. We used to in this country really believe that people should sort this question out for themselves, and what we learned in the 1930s was that when this privately determined system breaks down,

02:37:14.360 - 02:37:49.360

the government has to get involved because it has massive negative effects on the economy, so we have the FDIC, and if any of these private arrangements becomes big at scale and it breaks down, the government has to get involved. So, we better know what that means and what the back stops are if we're moving away from what we have now, which is not a great system, but it does have some robustness on points like that. Like, is the retail investor protected, and do they feel secure? Even in the height of the crisis in 2008, retail depositors did not run, did not try to withdraw their deposits from the banks, which is a remarkable achievement.

02:37:49.360 - 02:38:18.360

Perhaps to punctuate the point, we haven't made it actually as a panel, so perhaps to say it out loud. As we move from the old paradigm to the potentially new paradigm, we're going to have to represent fiat currency on this system somehow, and that fiat currency is going to have to be distributed and exchanged in a way that we are all used to having certain risks and certain custodial qualities that we need. That doesn't go away, and so we still have to solve for those problems.

02:38:18.360 - 02:38:49.360

So, central banks, as Simon rightly points out, are actually thinking about this, regulators are thinking about this, and the only way that we get there is if we all kind of work as one ecosystem, because otherwise it ain't going to fly. One thing that's interesting is that FDIC was deposits, and now there aren't any, but you're still going to have to have the same ability for the government to stand behind this. So, D will stand for digital in some way, and we can work out the other three letters. I think the whole thing sounds like Schrodinger's cashed to me.

02:38:49.360 - 02:39:02.360

B for bits. Yes. We've run over it because it's been such an interesting session, but why don't we thank this panel now for... Thank you.

02:39:02.360 - 02:39:29.360

And we now move to the public comment portion of the program. I believe there is just one public comment. Maxine Savitz will preside. Thank you, John. As Dr. Holdren mentioned in the beginning of the meeting, the PCS has several studies underway, and one of these has to do with how science and technology can be used to ensure the safety of our drinking water.

02:39:29.360 - 02:40:14.360

And the one public comment that we have today is from Dr. William Hersey. I know he will be settling in in a minute. Welcome, Dr. Hersey. He's the Science and Regulatory Affairs Advisor to the American Environmental Health Studies Project Flash Fluoride Action Network, and he's going to be talking about lead in drinking water and increases in children's blood levels. And as you know, we have a two-minute rule, and we'll let you know in 90 seconds.

02:40:14.360 - 02:40:34.360

Thank you very much. I'm also a retired senior scientist at EPA headquarters here in DC, and a charter member and past president of the Professionals Labor Union at EPA. And this issue came to my attention while serving as a union officer. Honorable Councilmembers, thanks for the opportunity to bring to your attention the material.

02:40:34.360 - 02:41:17.360

It's germane to lead in drinking water issue, which is of considerable interest to Congress, public health professionals, and the public at large. It was a recent article by Brady Dennis in the Washington Post that addressed this issue, focusing on problems with the EPA lead and copper rule. But unmentioned in this otherwise comprehensive piece is the relationship between lead leaching into drinking water and the chemical used in 90% of fluoridated water systems, which is fluorosilicic acid. This relationship has been known by EPA, local government officials, and others in the public health community for over a decade, but it has not been acknowledged nor acted upon in spite of its significance.

02:41:17.360 - 02:42:13.360

I can't provide a detailed treatment of this subject in a brief time I have today, but one is available written by Michael Connit and is at the URL that appears on my written comments. A peer-reviewed publication cited below and discussed at length in the Connit article have pointed out that fluorosilicic acid is a potent leacher of lead from leaded brass plumbing fixtures and lead service lines. And that its use in fluoridation systems increases blood lead levels in children compared with an alternative fluoridating agent, sodium fluoride, or in children drinking unfluoridated water. Given the widespread existence of lead service lines in communities using fluorosilicic acid, it would be a simple expedient for lowering lead leaching and children's blood lead levels to stop adding that particular chemical to drinking water, whether or not anti-corrosion measures such as addition of zinc phosphate are in place.

02:42:13.360 - 02:43:04.360

After all, lead phosphate particles can and do release from the deposited solids in these service lines and find their way into drinking water at the tap. It's better not to have lead leached from the metal in the first place and be happy to answer any questions. Thank you and you have provided some references. Thank you very much. That brings our program to a close. Again, thanks to the PCAST members, thanks to the panelists, and our one public commenter, and thanks to the audience, both in the room and over the web. And of course, thanks as always to our able PCAST staff, including particularly the PCAST secretariat that organizes these meetings. We are adjourned.