00:00:00.000 - 00:00:09.680

Okay, we should come to order. Welcome everybody. Welcome to the members of PCAST, to the OSTP staff who have joined us

00:00:09.680 - 00:00:22.000

and to members of the wider community of folks interested in science, technology and innovation who have joined us both in the room and for the live stream on the web.

00:00:22.000 - 00:00:32.640

Very happy to have you all here. As I think everybody knows, this is the last meeting of the full PCAST of the Obama administration.

00:00:32.640 - 00:00:44.560

We will be reviewing a little later in the morning the whole history of PCAST, the number of times we've met, the number of reports we've produced,

00:00:44.560 - 00:00:59.200

some discussion of impacts we've had, some thoughts about the challenges ahead. But for now, aside from those words of welcome, I want to turn it over to my co-chair, Eric Lander,

00:00:59.200 - 00:01:15.160

for any initial comments he might want to make. And then we will launch right into a discussion of some new reports that we have prepared and are ready for the vote of the group

00:01:15.160 - 00:01:26.120

as to whether they are approved. Eric. I want to use this chance to express my deep thanks to all the members of PCAST.

00:01:26.120 - 00:01:38.240

We'll talk, as John says, a little bit later in this meeting about the bigger picture of what PCAST has done over the last eight years. But I want to say to all of the members of PCAST

00:01:38.240 - 00:01:51.600

how much we appreciate everyone's service, their incredible hard work, and the fact that even up to this last meeting, people are dedicated to working on bringing

00:01:51.600 - 00:02:01.000

final studies to closure. It is a remarkable group. And I want to say thank you. I also want to make a point to say thank you to our staff.

00:02:01.000 - 00:02:13.120

Ashley and Jen Michael, Ashley Preetith and Jen Michael, have continued to support this group. And we have been blessed with an amazing staff over the course of eight years.

00:02:13.120 - 00:02:22.520

And we could not possibly have done this without you and without our colleagues in OSTP. So I want to start the meeting just recognizing how much everybody has put into it.

00:02:22.520 - 00:02:29.240

And we'll turn to our agenda and we'll have a chance to return to this point. But thank you to everybody. And thank you to everybody on the web

00:02:29.240 - 00:02:41.520

who has been watching, to people who've been attending, to all the people who will watch someday in the future. Because it's all archived if it turns out that folks have an interest in looking back

00:02:41.520 - 00:02:49.760

at what was accomplished and discussed over this period. We've created something of a record of things that people can turn to. So thank you.

00:02:51.200 - 00:03:03.640

Thank you, Eric. We have no public comment for this meeting. We had decided we would have the public comment session at the beginning, but there are not any scheduled.

00:03:03.640 - 00:03:19.520

So we will move straight to a presentation and discussion of the first study that's up for a vote. And that is the study on semiconductor challenges. That, the working group that produced that report

00:03:19.560 - 00:03:36.760

was co-chaired by Paul Otellini, the former CEO of Intel, and by me. Paul was not able to join us for this meeting. But another real expert in this domain, which I am not,

00:03:37.120 - 00:03:52.080

Craig Mundie, PCAST member, was an extremely active member of that working group and had a lot to do with on relatively short notice, our ability to drive that study to a successful conclusion.

00:03:52.080 - 00:04:03.760

So I'm going to ask Craig to present it this morning. So Craig, the floor is yours. Thanks, John. So let me begin with a little context

00:04:03.760 - 00:04:15.520

and motivation for the study. Obviously semiconductors are a key component of the life and world that we live in today. And the United States has, for many decades,

00:04:15.560 - 00:04:29.400

been a leader in the development of that. At various times in the past, the country has faced potential losses of leadership in these areas, but through a variety of strategies,

00:04:29.400 - 00:04:47.480

always managed to recover and remain in an important position even as the world has changed substantially. But right now we face two significant problems. One is that the natural technological evolution

00:04:47.800 - 00:05:08.360

as sees some slowing in the relatively near future, things people call Moore's Law, the famous doubling of chip densities and attendant performance benefits every 18 months or so

00:05:08.480 - 00:05:17.560

is starting to slow down. The cost of building these fabrication environments is getting very, very high. But we also see other countries,

00:05:17.560 - 00:05:31.120

notably China, who recognize the strategic importance of this and also recognize that China, for example, will be a substantial fraction of the total world market for semiconductor consumption

00:05:31.120 - 00:05:43.120

in the course of the next 10 or 20 years. And so they've embarked on a strong industrial policy targeting this industry and with a substantial investment toward that.

00:05:43.120 - 00:05:59.000

And so the goal of this study was to determine what the United States should do to try to ensure that its leadership in this key area remains strong. The slides here just show the representation on this committee

00:06:02.200 - 00:06:12.680

and included many key people who have been involved in the semiconductor industry for a long time directly, as well as economists and other people

00:06:12.680 - 00:06:30.480

who understand this from a national security standpoint. As I said, the two biggest issues if you summarize them are the change in the technological landscape and the competition that's gonna come from China.

00:06:30.480 - 00:06:40.680

As we thought about the study, we ended up with six guiding principles. Perhaps the most important is that you cannot win this by playing defense only.

00:06:40.680 - 00:06:54.840

And in fact, when the US has prevailed in previous challenge periods, it did so because either by strategy or serendipity, we evolved what was important in semiconductors

00:06:54.840 - 00:07:05.600

at a time that other countries were investing in the previous concepts. And so our goal here in the report is to encourage the country and the private sector

00:07:05.600 - 00:07:14.320

to think collectively about doing that with real intent this time. We wanna focus principally on leading edge semiconductor technologies.

00:07:14.320 - 00:07:25.960

This is another substantial variance from what we've seen in the past, in that today the diversity of use of semiconductors, which ranges from the smallest of small things,

00:07:25.960 - 00:07:37.720

the internet of things as it's emerging, to the very largest computers that we know, produce both an opportunity and a requirement for a very different approach in many cases

00:07:37.720 - 00:07:53.440

to what semiconductors provide. And so with that comes an opportunity to bring a larger part of the technological capacity of the country to bear on advancing in this area.

00:07:55.160 - 00:08:08.600

We do think that the US has strengths that don't yet exist elsewhere in the world. Including in China. And so rather than doing what might be a knee-jerk reaction,

00:08:08.600 - 00:08:16.040

which is to try to think, well, if China has an industrial policy, should the US suddenly decide it needs to have an industrial policy?

00:08:16.040 - 00:08:26.600

And that, of course, is not something that has been favored conceptually for this or any other area in the United States. And similarly now we think that

00:08:26.600 - 00:08:41.360

that would be the wrong approach. We do expect that as we begin to take steps to ensure an equitable playing field as China makes these policy-based investments,

00:08:41.360 - 00:08:52.320

and we begin to see the United States respond to those, that there will be obvious responses. But we think that those have to be well considered and measured on both sides.

00:08:53.320 - 00:09:08.480

And it's clear that while China will make advances, a reflexive response to the progress that they do make, if it diminishes our ability to advance toward a new concept would probably not be the best strategy.

00:09:08.480 - 00:09:19.800

And so we have to make sure that we maintain balance on that. And of course we want to enforce trade and investment rules such that where there are norms and legal obligations

00:09:19.800 - 00:09:31.880

on both sides and with other countries that we maintain our focus on supporting those. So in order to do that, we came up with a set of recommendations.

00:09:31.880 - 00:09:45.880

The first one was to in fact find a way to bring a set of industrial expertise and make it available to the government so that as they face questions about strategy

00:09:45.880 - 00:09:56.320

or action on the international stage, that they can be well informed because it's a complex technological area. It does continue to move quickly.

00:09:56.320 - 00:10:17.600

And so we think that we have to build that capability as well. Next thing that we want to talk about is really this question of sort of maintaining the balance. And I'll say the level playing field, if you will,

00:10:17.680 - 00:10:28.800

on an international basis. The US has a range of tools at its disposal when it feels that there's a threat, an unnatural threat of some kind.

00:10:28.800 - 00:10:40.160

And so things like the CFIIS mechanism for controlling investment when it affects national security, the long history of the potential for export controls,

00:10:40.160 - 00:10:53.480

some of which are actually in effect today on certain key manufacturing equipment. These things all remain in place. But as China in particular and potentially other countries

00:10:53.480 - 00:11:07.560

enact measures to try to improve their own lot relative to this critical and strategic industry, we will have to be cognizant that those policies and tools may need to be adjusted

00:11:07.560 - 00:11:20.160

in order to ensure that they're effective in the current geopolitical and technical environments. Toward that end, we think it's gonna be important to boost transparency of what's happening globally

00:11:20.160 - 00:11:29.600

with respect to these activities. We think it's gonna be important to bring that together and monitor it and talk about it. We think it's gonna be very important

00:11:29.600 - 00:11:42.800

to maintain a active dialogue with China. What they're doing is completely in the open from a policy and investment point of view. And so we need to make sure that we understand

00:11:43.040 - 00:11:56.800

and that they understand what we think the issues are around this to the extent that the tools that we have need to be reshaped. We need to be sure that that's done in a timely way.

00:11:58.360 - 00:12:14.280

The third is also to recognize that the US has a number of natural allies, other countries and companies essentially that work on a global basis in this domain and that they too would be concerned

00:12:14.280 - 00:12:28.840

if they saw an unnatural consolidation or sort of non-trade based ways of tipping the playing field in some way. And so we think that it'll be very important

00:12:28.840 - 00:12:44.680

that diplomacy is applied here with respect to the other countries who share a natural concern and that we use that to boost the transparency and make sure that we have common cause

00:12:44.680 - 00:12:59.640

with respect to these kind of controls. Next area that we focused on is trying to make sure that we create an optimal environment in the United States for the companies that want to invest

00:12:59.640 - 00:13:14.180

to build this next generation of capabilities. And the US has of course long been a favorable place for this to happen and we have a long history here. But many of the companies that are the world leaders

00:13:14.180 - 00:13:29.640

in both the semiconductor industry itself and the related computing areas are very global companies. And at this point, the global market, in fact the market for their products

00:13:29.640 - 00:13:42.660

is larger outside the United States in most cases than it is in the United States. And so there's a natural tension and some concern that if the US in any way is just even less favorable

00:13:42.920 - 00:13:56.120

let alone hostile to these kind of investments, then there will be a lot of implicit pressure for the countries to be neutral or less concerned about this evolution in the market.

00:13:57.040 - 00:14:10.960

But we also know that if policy is leading toward consolidation, the companies would also find themselves at risk in the long term. And so I think that there's certainly area

00:14:10.960 - 00:14:25.720

for our collaboration around these issues. So we made a number of recommendations. One not new as a general concept for this group, PCAST is the importance of STEM education

00:14:25.720 - 00:14:41.920

and the development of the talent pool. Immigration reform continues to be a key issue in this area. Lots of the academic research work and other things that underpin this

00:14:41.920 - 00:14:51.120

is done outside the United States, particularly in Europe as well as in Asia. And so we have to pay close attention to the talent pipeline.

00:14:52.120 - 00:15:08.020

We want to continue to invest in pre-competitive research. That is part of the fuel that makes all this run. But notably, as you'll see in the next section, we believe that the window to create this next generation

00:15:08.020 - 00:15:21.780

of semiconductors is a fairly narrow one, perhaps the next 10 years being the principal time of opportunity. And as such, we can't begin to look for new research

00:15:21.780 - 00:15:36.900

to do this, the time horizon would be too long. And so we have to really continue to invest for the future, but the principal focus as this report will show has to be to harvest a lot of the fundamental science

00:15:36.900 - 00:15:48.260

that has been done in the United States and around the world and begin to invest, to bring it to market and recognize that to do that, we have to overcome a number of implicit risks

00:15:48.260 - 00:16:00.300

that would otherwise deter that. We think there are other specific things that could be done, maybe some changes from a tax reform point of view, strengthening R&D tax credits,

00:16:00.300 - 00:16:11.120

all of these just basically create a more favorable environment for the large investments, pretty capital investments that are required in order to develop these new capabilities.

00:16:12.060 - 00:16:25.580

There are some areas where the semiconductor companies in particular want some attention paid, both at the state level and potentially at the EPA federal level.

00:16:25.580 - 00:16:38.820

And that relates to the permitting processes associated with these complicated fabs that produce the products. These processes have a lot of complex chemical properties.

00:16:38.820 - 00:16:53.860

Those properties have resulted in a lot of regulatory scrutiny in the construction of the fabs. And that permitting process is becoming a longer and longer part of the total cycle time

00:16:53.860 - 00:17:05.540

as the technological capability to build the fabs has actually shortened their construction time and time to put them into operation. And so time is money in this case

00:17:05.540 - 00:17:20.540

and the permitting process needs to be at least made as expedited as possible. Finally, and perhaps the real key in this report is a recommended strategy for how the country

00:17:20.540 - 00:17:36.060

should approach creating the next big thing in semiconductor or related component technologies. Normally there might be a temptation to just try to respond to the policy action

00:17:36.060 - 00:17:45.260

or the investment action on the part of China or other countries. But our strong view in this group, I think the unanimous view was that

00:17:45.260 - 00:17:56.700

unless the country can establish some very broad goals that tend to move us in a direction that is not incremental improvement of the current production capabilities,

00:17:56.700 - 00:18:10.180

that it's unlikely that the US would sustain the advantages that have accrued from being a leader in this domain. And so the report goes on to propose a series of moonshots.

00:18:11.180 - 00:18:26.740

And but here the goal is not a moonshot specifically to solve the problem that you find on the title of the moonshot. The goal is to use these important problems

00:18:26.740 - 00:18:37.300

as a way to galvanize collective action between the public sector and the private sector and across a range of domains and technologies that might not otherwise be brought to bear.

00:18:37.300 - 00:18:45.940

And so and importantly, I mean, some, there are activities in many of these cases where people would say, well, we're already trying to do some of that,

00:18:45.940 - 00:18:58.300

but they would all be being done within a problem domain using the currently available technologies. And so here, you know, the goal is actually the reciprocal, which is to look for a radical approach

00:18:58.300 - 00:19:12.020

that if achieved would in fact be a benefit, but where the byproduct is intent is the intent with the intention to produce breakthroughs in the underlying computing and semiconductor technology.

00:19:12.020 - 00:19:21.100

And this will require a mix of private and public participation. And if you read the report in detail, you'll see that we actually identify

00:19:21.100 - 00:19:32.080

some problem domains where businesses are likely to invest heavily to pursue them anyway. But there's a fairly substantial set of areas that are important to the overall economy,

00:19:32.080 - 00:19:41.380

but where the computing and semiconductor industries would not naturally take a leadership role in trying to prosecute them. And so some of these things are gonna require

00:19:41.380 - 00:19:54.980

more government investment and activity to bootstrap them while others might be, the government has to ride along on what will otherwise be initiatives at scale by industry itself.

00:19:54.980 - 00:20:08.740

So we wrote three sample moon shots. In the appendix, we identify a very broad inventory of what we call component technology vectors that we think are all harvestable

00:20:08.740 - 00:20:23.060

within this 10-year window. And each of which represents a potential tool with which to attack a moonshot class problem. But to show how we think that this mechanism would work,

00:20:23.060 - 00:20:34.980

we actually wrote hypothetically three moon shots in order to guide this process. The first was the development of a zero-day bio threat detection network.

00:20:35.980 - 00:20:47.740

PCAST recently did a report on bio hazards. And one of the things that's of course recognized is that downstream with the emergence of synthetic biology and other things,

00:20:47.740 - 00:21:01.700

it's hard to predict what and from what quarter we might see a biological threat, whether with mal intent or otherwise. And so it's a very difficult problem,

00:21:01.700 - 00:21:14.300

in general one that we don't know how to solve, but it requires a new class of sensing and a new class of computing in order to be able to do this because it would be in a defense environment in a world

00:21:14.300 - 00:21:24.380

where by this time, we might expect to see the emergence of quantum computation. You know, it requires that we bring a whole new security model forward in the solution as well.

00:21:24.380 - 00:21:37.660

So this basically is an example of a problem that would be valuable to solve. We either don't have a good solution to or would like a much better solution to this,

00:21:37.660 - 00:21:49.900

and which would require advances in many of these areas in order to have a complete offering. The second one, well, so what did we want to see?

00:21:51.260 - 00:22:03.540

Not sure how that one got out of order maybe. But so we wanted to, I guess that was just one example. There were two others, and one was essentially a focus on

00:22:03.540 - 00:22:15.060

high resolution weather modeling, one kilometer global resolution. This is something that currently is unachievable, and the current approaches in exascale computation,

00:22:15.060 - 00:22:28.620

you know, basically require power levels that are quite limiting in terms of when and how many of these we might be able to deploy. So in the proposal we look at,

00:22:28.620 - 00:22:37.980

much more aggressive way of approaching this, using new methods and techniques, and notably not focused on using a classic CMOS based technological system.

00:22:38.940 - 00:22:55.460

And the third is basically one to develop a new model of high scale zero carbon energy production. Another important goal, and one where we think,

00:22:55.460 - 00:23:07.660

without breakthroughs in both materials and chemistry, we probably won't get there. And so the goal suggests that we should develop a quantum computation capability commercially

00:23:07.660 - 00:23:19.060

that could be deployed against the materials and chemistry problems, and from that potentially bootstrap a different class of energy solution.

00:23:19.060 - 00:23:33.140

So we then suggested that we need to see these things formalized and orchestrated. Perhaps there should be other moon shots in order to get a complete cover set

00:23:33.140 - 00:23:41.220

of the important technologies. And so we recommend that there should be a subcommittee under the NSTC that would be put in place

00:23:41.220 - 00:23:55.260

in order to administer this kind of activity. So that's a quick summary of what's in the report, and I'm happy to take questions. Thanks very much, Craig, for that excellent summary.

00:23:57.100 - 00:24:07.780

Floor is now open to PCAST members with the usual protocol of raising your flag when you would like to comment, and Bill Press is the first flag.

00:24:10.140 - 00:24:20.300

Thanks, Craig. I think this is a very important report. I think it has a heritage that comes substantially out of real industry experts as well as academics,

00:24:20.300 - 00:24:33.180

and I hope it's very influential. My question is on the moon shots, I'm visualizing it as kind of column A and column B, and column A is what are the important problems

00:24:33.180 - 00:24:47.380

that we think that different kinds of computing, more advanced computing, will be able to solve, and column B is this list of technology vectors that are currently underexploited and could be exploited.

00:24:48.460 - 00:25:04.980

My question is in deciding which vectors go with which problems, are you viewing that as prescriptive? Is that the final answer, or if a corporation or an academic research group had an idea

00:25:04.980 - 00:25:20.380

that they could mix and match from those two columns differently, is that as good a solution? Yeah, as we said, we offered these as an example, more to stimulate thinking and get people recognizing,

00:25:20.460 - 00:25:34.860

much like the original, with the man on the moon moonshot, nobody knew exactly how we were gonna do it when Kennedy's got up and said, let's do that. But the country brought its assets together

00:25:34.860 - 00:25:49.580

and achieved that in a 10-year period. So I think that the moonshot term applies in that regard. Here we thought that there were a set of important things, some of them, for example, in recognizing

00:25:49.580 - 00:26:01.740

we want more and more interaction in the biologics area. In some sense, that's not classic semiconductors, and yet we wanna see a hybridization. So we constructed the three examples

00:26:01.740 - 00:26:10.940

to try to show how you start to get a cover set of things, but we're not totally prescriptive or don't intend to be totally prescriptive, and in fact, we published a whole set

00:26:10.980 - 00:26:22.940

of technology vectors that came out of the inventory, largely to encourage either more moonshots to be developed or people to take a different approach. Chad Morgan.

00:26:22.940 - 00:26:33.940

Craig, I think that was a fantastic summary. Looking at the moonshots, I think they're just a beautiful compilation of the challenge here and the opportunity

00:26:33.940 - 00:26:47.180

because it is a nice example of the importance of convergence in creating a competitive advantage, bringing fields into the semiconductor arena that traditionally are not impacting them

00:26:47.180 - 00:26:57.740

to create much higher value items. The one moonshot you described in detail, it seems that the groups of folks that would be interested could be expanded.

00:26:57.740 - 00:27:10.620

I'm curious why IARPA and DHS, for example, wouldn't be called out there as well because it seems like a very obvious set of agencies that would take an interest

00:27:10.620 - 00:27:22.620

in the one that you described in detail. I think you're right. I mean, we, both in terms of how we think you approach this and which government agencies might be interested or involved,

00:27:23.780 - 00:27:30.060

that's all up for grabs. It's why we said these are examples, these aren't the specific answer and why there should be a committee

00:27:30.060 - 00:27:40.300

who looks at this in more depth and with a real understanding in the next administration, in particular of what the areas of focus might be agency by agency.

00:27:40.300 - 00:27:53.940

But we did think it would be important that there be an obvious lead agency or a couple of lead agencies whose natural mission requirement would focus in these areas

00:27:53.940 - 00:28:05.700

and we tried to identify those, but I certainly wouldn't deem that as comprehensive or that it should exclude any of the other agencies who have a natural interest in participation.

00:28:08.740 - 00:28:22.940

Other questions? Other questions or comments for Craig. I know a number of the other members of PCAS were themselves paying quite close attention to this study,

00:28:22.940 - 00:28:40.820

but it seems that Craig's summary has satisfied all interests and concerns. That being so, I would like to call for a vote as to PCAS's approval of the report on semiconductors

00:28:41.620 - 00:28:55.540

as you will subject to final minor edits. All those in favor, please raise your hand. All those opposed, those abstaining. The ayes have it unanimously.

00:28:55.540 - 00:29:12.780

The semiconductor report is approved and will be transmitted in due course to the president and posted on the PCAS webpage. Thank you again, Craig, and thanks by remote control

00:29:12.780 - 00:29:26.700

to all of the members of the working group who included, as Craig noted, an extraordinary array of both industry and academic experts on the array of issues

00:29:26.700 - 00:29:41.180

raised by these semiconductor challenges. Let us move now to the National Nanotechnology Initiative Update and Vote, and Michael McQuade, I believe,

00:29:41.180 - 00:29:48.940

will be leading that discussion. Michael, the floor is yours. Please remember to press your microphone. Thank you very much.

00:29:48.940 - 00:29:56.780

I'll make just, this is pretty short. I'll make some brief comments and then Mark maybe give you a heads up to prepare a little bit too. This is the sixth review

00:29:56.820 - 00:30:09.940

of the National Nanotechnology Initiative. Just to provide a little background, NNI has been instantiated since 2001, and in that time the US government has spent

00:30:09.940 - 00:30:20.260

approximately $24 billion on the National Nanotechnology Initiative. The peak funding was about $1.8 or $1.9 billion a couple of years ago.

00:30:20.260 - 00:30:33.900

It's about $1.5 billion in the current fiscal year. And just then to come from the other direction, the latest estimates from commercial research reports are there are in excess of a trillion dollars

00:30:33.900 - 00:30:43.380

of nanotechnology enabled products and services in the marketplace today. So it is substantial investment we've made and it has had substantial returns.

00:30:45.340 - 00:30:53.980

This is the sixth review. We and NRC are both currently statutorily required to review the NNI on a periodic basis.

00:30:53.980 - 00:31:05.420

So we have done that every two to three years as has the NRC. And we should say that in the recent competitive act funding of 2016,

00:31:05.420 - 00:31:14.420

based on the success of both the program and the review process, we are very pleased that Congress has changed the review process

00:31:14.420 - 00:31:27.260

from both agencies to every four years. And one of our recommendations will be how to offset that. There are two primary recommendations in this report. The report is quite short based

00:31:27.260 - 00:31:37.340

on the comprehensive review that was done in late 2014 by PCAST and the recently released NRC report. The NRC report in particular

00:31:37.340 - 00:31:50.000

continued to focus on many of the same issues that had been highlighted in our 2014 report. In particular, for those of you who will recall in 2014, one of our major reports or recommendations

00:31:50.000 - 00:32:06.840

was to initiate the structure of Grand Challenges, a very similar discussion to what we have just been having. Challenge problems that to first order are of significance. They have scale and a belief that nanotechnology

00:32:06.840 - 00:32:21.160

of some kind can be an important mechanism in solving those Grand Challenges. We were very pleased that under Lloyd Whitman's leadership, the NNI community came forward last year

00:32:21.160 - 00:32:31.280

with its first Grand Challenge, Nano Inspired Grand Challenge, related to highly efficient computation at the scale that would reproduce the energy efficiency

00:32:31.280 - 00:32:38.500

of a computational system related to brain capability in humans. So a tie to both the nano initiative and the brain initiative.

00:32:38.500 - 00:32:50.160

Our first recommendation in the report this year is that in keeping with the impact of Grand Challenges and the success of the initial Grand Challenges that we recommend that NSET continue that process

00:32:50.160 - 00:33:01.080

and announce in the next year to year and a half two more Grand Challenges to be owned and operated by the National Nanotechnology Initiative. So that's our primary technical recommendation.

00:33:01.080 - 00:33:10.720

We further than endorse the recommendations again from the 2014 report, knowing that nothing substantially has changed in that time period.

00:33:10.720 - 00:33:19.720

Those recommendations are still valid and have been endorsed by the NRC. And then finally, we recommend that the next comprehensive review

00:33:19.720 - 00:33:33.640

by PCAST be done in 2018 and that subsequently PCAST doing its review and NRC doing its review are offset by every two years. So we are not doing a review every year

00:33:33.640 - 00:33:39.560

of the program any longer. It's being done every two years and each agency that's responsible would be on a four year clock cycle.

00:33:40.660 - 00:33:48.520

That's the sum in total of the report. Now I'd like to turn it over to Mark, maybe for some additional comments. Mike, thank you.

00:33:48.520 - 00:34:00.160

And I particularly want to thank and praise Mike for his leadership here on this report and particularly call Ashley Prita for her amazing work, not just in this cycle,

00:34:00.160 - 00:34:13.600

but in the cycle before when this report came together. Is it, because this is really a continuation of the work that we did in 2014 to review and a look at what's happened in the last two years

00:34:13.640 - 00:34:25.640

since that happened. And people believe that these reviews are fairly routine, essentially report cards, but I would say that the review in 2014

00:34:25.640 - 00:34:40.680

was clearly not standard. If anything, the NNI had gone through a period where it was primarily a coordinating body. And by 2014, many people here in the government

00:34:40.800 - 00:34:57.080

outside wondered if the NNI had to run its course. And we looked at it, we looked at that seriously as Maxine challenged us, but we also looked at it another way,

00:34:57.080 - 00:35:10.600

which is that nanotechnology is really, just in its very early innings, it is not a specific technology. It's not tied to any particular industry

00:35:10.600 - 00:35:24.440

or discipline or any agency, any agency silo. It's clearly a revolution in how we work with atoms. And so we're at those very early innings and we saw a great opportunity

00:35:24.440 - 00:35:36.720

to move from coordination to leadership. And that leadership took on this concept of the Grand Challenges. And if anything, Lloyd has redeemed our thoughts there

00:35:38.200 - 00:35:48.040

by doing such an exemplar job in bringing this forward. And if anything, it's our belief that, and I'll speak for all of us on this,

00:35:48.040 - 00:36:00.720

that the NNI probably is more important now because of that than it has been in its last five cycles of viewing this coordination. So we've moved to this place now

00:36:00.720 - 00:36:12.760

where if anything, we're extremely enthusiastic about the importance of the NNI. And perhaps that's the more important thing that we're voting on today rather than this review

00:36:12.760 - 00:36:26.040

is our continued belief in the importance of the NNI. And so we're very honored to have as part of PCAST, by the way, one of the world experts in Chad Merkin from Northwestern.

00:36:26.040 - 00:36:36.240

But I can tell you that our leading universities right now are seeing this as the starting point. I know this is the number one thing that MIT is doing today,

00:36:36.240 - 00:36:50.000

building a 200,000 foot building in the center of campus. And it's the number one thing that they're funding today. So I think we're seeing just the very beginning of something that can change the human health

00:36:50.000 - 00:36:57.440

for the good and change the health of the planet for the good. And so very enthusiastically support the recommendations here.

00:36:58.440 - 00:37:16.040

I just want to add, if you would want to thank Michael and Mark. And this is an initiative that as Michael said, began 2001, 2000, you can actually go to the Clinton Library and see the initial speech that he gave at Caltech

00:37:16.040 - 00:37:27.120

to introduce this. And this has been a really bipartisan program both from the Congress and the administration of carrying it through starting with five agencies

00:37:27.120 - 00:37:36.480

and expanding now to 20 some agencies. So it's been part of their missions but also with the private sector. But through this has been very important

00:37:36.480 - 00:37:49.920

that the federal government's role continue in the fundamental research area, which it has done, and particularly as it moves through applied programs and also to ensure US leadership.

00:37:49.920 - 00:38:01.960

So it's been sort of exciting to see products on the market and nano has become a common word that everybody uses, may not correctly, but it's, and I think some of the biggest advances

00:38:01.960 - 00:38:17.720

in chat has this, have been in the health fair area that we're not necessarily anticipated at the beginning, but I think it's really a tribute to all the researchers in the industry itself and also the government

00:38:17.720 - 00:38:32.520

in sustaining a program that can lead from research to products that we become related. Good, Chad? Yeah, kudos to putting together a great report

00:38:32.520 - 00:38:48.040

and great, I think, summary and analysis of the situation because I think everything you've said is right on the mark. As we think about going forward, I think you guys have done a beautiful job

00:38:48.400 - 00:39:01.960

of emphasizing the importance of creating grand challenges to create a convergence of all these capabilities that can impact almost every field, and we've seen it really impact almost every agency

00:39:01.960 - 00:39:13.720

and every group of scientists that are working under the purview of those agencies. How do we plan for the future and the next set of grand challenges?

00:39:13.720 - 00:39:27.440

Is there a process in place? Do we have ideas on how the next set of grand challenges are going to be selected and are there some thoughts in that regard?

00:39:27.440 - 00:39:34.000

Yeah, I think I'll speak for Lloyd, who's sitting right behind you. I think the project that was used to select the first grand challenge

00:39:34.000 - 00:39:44.240

was in many ways exactly what we envisioned when we did the last report. So if you recall in that report, we focused, and again, I'm gonna draw an analogy to what Craig just said,

00:39:44.280 - 00:39:53.560

we focused on the grand challenge, but we also focused on the process of bringing the community together to instantiate and decide what the grand challenge is.

00:39:53.560 - 00:40:04.000

So Lloyd ran this as part of the council. There were significant public meetings that were held. There was a request for proposals. And I'm sure if I get there,

00:40:04.000 - 00:40:14.120

whether it was an RFP or an FOA or whatever the mechanism that was used, but there was a very strong engagement with the community. And then the council ultimately selected

00:40:14.120 - 00:40:24.240

what this particular grand challenge was. We, I think, would envision the same process running again. And that's what we had suggested when we did the recommendations last time.

00:40:24.240 - 00:40:33.800

So I don't know, you're sitting in the back. I don't know if you want to say anything or Lloyd? Getting a lot of people into that. We're fighting today to face all that,

00:40:33.800 - 00:40:46.000

including the first game, sort of, talking to many, many people across the community about what has this now been done through the law of direct stakeholder views

00:40:46.000 - 00:40:57.040

and how people have lost their control. Basically, we can support it today, but that's not right for us. I'm going to write down some comments.

00:41:03.160 - 00:41:17.480

Seeing no other flags up, I think we can move to a vote on approving this latest update of our review of the National Nanotechnology Initiative,

00:41:17.480 - 00:41:34.360

taking into account the comments that were made, that we are really voting in a sense on the culmination of a longer and broader effort to focus and advance this particular initiative.

00:41:34.360 - 00:41:45.120

And again, before we vote, I would want to add my congratulations and thanks to the team, both the PCAST team and the extraordinary staff

00:41:45.120 - 00:42:01.520

who have worked to advance this initiative. So all those in favor of approving this installment and endorsing our long effort, hands up, opposed, abstained, Chad abstains,

00:42:01.520 - 00:42:13.640

I think because of his connections to an important part of the initiative. Not through any reluctance. No, as he has consistently on each nano report,

00:42:13.640 - 00:42:29.640

Chad has followed that policy and we respect that. So the report is approved and we move now to the third new piece of PCAST work to be discussed here

00:42:29.760 - 00:42:42.760

and that is the update or addendum to the Forensic Science Report that the group released in September. And for that, I will call on my PCAST co-chair

00:42:42.760 - 00:42:50.280

but also the chair of the Forensic Science Working Group, Eric Lander. Great, thank you, John. So I'll give a brief update.

00:42:50.280 - 00:43:05.480

PCAST has produced an addendum that we bring forward for the approval of the group today to the report we had just to remind us without revisiting the entirety of 174 page report.

00:43:05.800 - 00:43:17.040

The basic point of the report was that forensic methods in order to be considered reliable, which is the standard they must meet to be admissible in court,

00:43:18.360 - 00:43:28.840

have to have been tested empirically. That's pretty much a summary of the 174 pages. If you haven't gotten empirical tests that something is reliable, you can't know it's reliable.

00:43:28.880 - 00:43:50.000

So we undertook a very extensive review of the literature. We looked at more than 2000 papers to cover a wide variety of fields, seven fields in all and found that in some cases there was really convincing

00:43:50.400 - 00:44:02.560

evidence that people had done empirical studies, demonstrated reliability, could actually measure reliability if you can't put at least some rough number to it, you haven't measured it.

00:44:02.560 - 00:44:12.780

And in other cases at the extreme, there were no studies whatsoever that had ever been done that demonstrated reliability. So it's covered the whole spectrum there.

00:44:13.840 - 00:44:25.880

We recognize that within the field of forensics, there's a growing group of people who really see themselves as committed to the necessity of empirical studies and we applaud them

00:44:25.880 - 00:44:39.120

and we also strongly stand by the notion that it's not just a good idea, but it's a necessary idea to have empirical evidence. So we wrote this report, shared it broadly

00:44:39.120 - 00:44:52.080

and in response to the report, we got lots of feedback from many people in the community, but two that were particularly important pieces of feedback. Some groups asserted that empirical evidence

00:44:52.080 - 00:45:04.640

was indeed the best way to show something was reliable, but suggested that maybe you could show things were reliable without empirical evidence, although no one actually suggested how you might do that.

00:45:04.640 - 00:45:18.560

The other was that there was a suggestion, particularly from the Department of Justice, that perhaps PCAST had failed to note some studies that met our criteria for properly designed empirical studies

00:45:18.560 - 00:45:28.960

that demonstrated the reliability and scientific validity of some of the methods, because we are empiricists and scientists and technologists, we take such issues seriously

00:45:28.960 - 00:45:41.600

and so we wrote to a number of organizations, including the Department of Justice itself, asking if people could point us to actual empirical studies of the forensic methods

00:45:42.400 - 00:45:51.400

we had reviewed that met the criteria of being a properly designed study and that had not been considered in the course of the PCAST report.

00:45:51.400 - 00:46:01.120

We received replies back on all of that, which we've sifted through. We got a couple hundred papers that people pointed to as being of interest to look at.

00:46:01.120 - 00:46:14.240

We looked at all of those. NIST even suggested that to be really comprehensive. We look at Interpol's web-based resource of 8,000 forensic papers happily sorted by discipline,

00:46:14.240 - 00:46:27.840

so we did that as well. And we've written a short eight or nine page addendum to the report that we bring forward to PCAST today that basically points to a handful of conclusions.

00:46:27.840 - 00:46:42.160

Number one, that there is no way to demonstrate reliability without actually empirically testing reliability. You can't make a conclusion that something is reliable without trying it out.

00:46:43.160 - 00:46:56.000

That I have to underscore does not in any way diminish the importance of many other things. People point to the importance of professional organizations in forensic science and accreditation and certification.

00:46:56.000 - 00:47:09.360

Those are all really important things, but they don't in themselves say anything about reliability. People point to large bodies of peer-reviewed literature that show patterns, fingerprints, footprints, et cetera,

00:47:09.360 - 00:47:21.640

differ and that therefore you might hope to be able to identify things based on those differences. That's a very important body of work to know that it's worth trying to develop a method,

00:47:21.640 - 00:47:34.040

but the existence of differences between things doesn't mean that you have a reliable method for associating items with their sources that you can only tell by empirically doing it.

00:47:34.040 - 00:47:44.640

So on considering this we found that we could not imagine, I guess science is premised on the idea that conclusions that a method are reliable require empirical testing

00:47:44.840 - 00:47:56.040

and that no one could provide us any suggestion of any way that one could know that something was reliable without empirical testing. So we've discussed that in the summary.

00:47:56.040 - 00:48:08.400

We also reported on the various possibilities of were there any studies that had been missed that were actual empirical demonstrations of the reliability of any of the methods

00:48:08.400 - 00:48:19.240

which we had felt short of or completely missed that standard and the short answer is no. We could find them. I note in particular that we had been moved to do it

00:48:19.240 - 00:48:31.080

because the Department of Justice had raised this possibility. In the end, the Department of Justice did not send us any such studies and we reached out particularly and spoke to the office

00:48:31.080 - 00:48:37.880

of the Deputy Attorney General and we were told that on reflection the department had no such studies to bring forward to PCAST.

00:48:37.880 - 00:48:50.640

And so we very much appreciate their raising the possibility and we were heartened to find that they too could not find and give us any studies that met PCAST standards and demonstrated reliability

00:48:50.640 - 00:49:00.640

but we felt it's important to add to the report by saying those things. And then finally, another point that was raised was with all of our focus on empirical studies

00:49:00.640 - 00:49:10.800

to develop to demonstrate reliability there was concern expressed that we shouldn't forget about the importance of other studies like ways to understand

00:49:10.800 - 00:49:23.840

why people make errors, not just measuring error rates but understanding the process by which people make errors or technology efforts to convert subjective methods to objective methods.

00:49:23.840 - 00:49:36.080

And there was concern that a focus on reliability alone might divert attention from those important efforts to advance forensic science. We have tremendous respect for those studies.

00:49:36.080 - 00:49:48.600

The first, they're called white box studies to see what's going on inside the black box and these efforts to create objective technologies and in no way would we want the absolute essentiality

00:49:48.600 - 00:49:57.520

of empirical studies to diminish from the other work that's going on. I don't think we think that there is a need for that. There should be funding for both

00:49:57.520 - 00:50:06.200

and it's a very important thing that we learn those things because that's how the technologies will continue to advance. So we've written a summary of that.

00:50:06.200 - 00:50:17.880

It stands as an addendum. We bring to PCAST to our longer report and that's pretty much my report on the subject. Thank you, Eric.

00:50:17.880 - 00:50:29.600

Do we have questions, comments? Jim Gates, who has himself been a key member of that working group and involved in these issues in other fora as well, Jim.

00:50:29.600 - 00:50:40.280

Thank you, John. And first of all, thank you, Eric, for the tremendous amount of work that this was completed on.

00:50:41.200 - 00:50:56.520

It was also, of course, we thank all of the people who work with you, like Tonya Simoncelli and your staff members and OSTP staff. I think the most important thing you have said very well

00:50:56.520 - 00:51:09.600

and since this is a public session, I think it bears some repeating. No less a person than Albert Einstein essentially said that Galileo was the father, not just of physics,

00:51:09.600 - 00:51:20.280

but all science. And the reason that Einstein ascribed this to Galileo because he, and this is his phrase, he said, Galileo drummed it into our heads

00:51:20.280 - 00:51:33.360

that science must be based on empirical observation and that until this idea was accepted, it would have been impossible for the Western birth of science to have occurred.

00:51:33.400 - 00:51:46.840

So this is an essential point about science, that it's not a faith-based belief system. It's a system that's grounded in the fact that we have to refer to nature in order

00:51:46.840 - 00:52:02.120

to be assured that we are doing science. And this also means, of course, that we are particularly aware of our limitations and that science gets codified in error bars

00:52:02.120 - 00:52:15.680

and level states of confidence and what have you. And the message that I think the report is attempting to deliver again to the forensic science community is if you are a science, then you must adhere

00:52:15.680 - 00:52:28.880

to this basic principle about what it means to be science. Like you, I have confidence that there are members of this community who understand this intrinsically and are pushing forward with it.

00:52:28.880 - 00:52:41.200

I recently, partly, as you well know, have become in the Defense Science Standards Board as a member of working with this community to make sure that we keep focus on this effort.

00:52:41.200 - 00:52:52.120

And I've met forensic scientists who are absolutely the people who will drive this forward. I think this report empowers and encourages them, so I think that's a very important function.

00:52:52.120 - 00:53:04.920

And the other thing that I would let it be known is that some of our previous discussion here within PCAST itself has resulted in ongoing discussions about moving some of the technology

00:53:04.920 - 00:53:19.400

from subjective observation to actually IT-based and modern computer technology to replace the subjectivity. So this report has already started discussions.

00:53:19.400 - 00:53:30.160

I commend you, as I said, for your leadership. I'm very happy to be part of the working group. And I will continue to be part of this other community as a consequence and continue to try and drive this message.

00:53:30.160 - 00:53:40.840

So thank you. Well, thank you for your continued service and the various capacities in this forensic community. Are there other questions or comments on the addendum

00:53:40.840 - 00:53:52.720

to the Forensic Science Report? I see none. And so again, with thanks to those who contributed, led by Eric Lander.

00:53:52.720 - 00:54:02.920

Very valuable contributions by staff. And I want to really call out Tanya and Diana Pankovich, who's no longer with us, and Ashley and others for... It's not easy.

00:54:02.920 - 00:54:13.400

And Jen, Michael, who's putting together all of these documents so that we'll have them up on the web in time. It's no easy feat to wrangle

00:54:13.400 - 00:54:22.320

so many different papers and things. So this was a team effort. I echo that and call for the vote. All those in favor of approving this addendum,

00:54:22.320 - 00:54:29.200

hands up please. All those opposed. Those abstaining. The ayes have it unanimously.

00:54:29.200 - 00:54:47.920

Again, congratulations to the team as well as thanks. We are now actually somewhat ahead of schedule. I think we can move rather than taking such an early break. I think we could move to the next topic,

00:54:47.920 - 00:55:02.040

which is PCAST accomplishments and vision for the future. And I believe we have some slides to start that discussion. I will make a few initial observations

00:55:02.040 - 00:55:11.760

based on those slides, and then we will open it up for a wider discussion. Oh, there it is. All right.

00:55:11.760 - 00:55:27.640

Good. So just a very quick recapitulation of the history of this operation. I should say that PCAST as an institution

00:55:27.640 - 00:55:39.720

goes back many, many years, really all the way back to the Eisenhower administration when it was called the President's Science Advisory Committee, PSAC, but had the same function as PCAST has today,

00:55:39.720 - 00:55:55.360

which is to support the President's Science and Technology Advisor and the OSTP in ensuring that the President gets the most objective and independent science and technology advice available and in the process of doing

00:55:55.360 - 00:56:09.880

so, that there is a mechanism that reaches out to the wider community of scientists, engineers, and innovators across academia, industry, civil society.

00:56:09.920 - 00:56:24.160

Because as all of the members know, and probably everyone in the room knows, the members of PCAST, other than the President's Science Advisor, which at the moment is myself, keep their day jobs.

00:56:24.160 - 00:56:41.000

And as a general matter, tend to have very demanding day jobs, but nonetheless contribute their time, their energy, and their intellects to ensuring the connectivity of the stream of advice that goes to the President

00:56:41.000 - 00:56:54.520

with that wider community, that advice reflects the insights of that broader community. President Obama announced the members of his PCAST at the annual meeting of the National Academy of Sciences

00:56:54.520 - 00:57:16.920

on April 27, 2009, just a few months after his inauguration in the first term. And the picture here shows the President meeting shortly thereafter with his PCAST in the Roosevelt Room.

00:57:16.920 - 00:57:29.880

We currently have 19 members on PCAST. A total of 25 people have served as PCAST members during this administration, 15 of them from the very beginning.

00:57:29.880 - 00:57:43.760

And these slides will be posted if some of the print is a little small for reading around the room. These slides will be posted, but this shows the members and the terms in which they served.

00:57:48.240 - 00:58:01.040

Today's meeting marks the 50th public meeting of PCAST over the course of the Obama administration, 45 in-person meetings at various locations, although I would say a majority of them

00:58:01.040 - 00:58:13.840

at the National Academy of Sciences, to which we are very grateful for the support and the assistance in PCAST work, provided by the use of the Academy's excellent facilities.

00:58:13.840 - 00:58:31.320

Five public teleconferences took place as well, all announced as required by law in the Federal Register. By the end of the administration, which is imminent, we will have published 39 PCAST reports,

00:58:31.360 - 00:58:47.760

37 of them unclassified, two classified, as noted here, the average page count is 80. I'm not sure that's a distinction, but it is fortunate that we have worked for a president

00:58:47.760 - 00:58:59.160

who is not only interested and enthusiastic about the application of science, technology, and innovation to national goals ranging from the economy to public health, to the environment,

00:58:59.200 - 00:59:12.640

to energy, to national and homeland security, but a president who is a voracious reader, and therefore the large average size of the PCAST reports has not been a deterrent

00:59:12.640 - 00:59:27.120

to the presidents absorbing their content and embracing a remarkable fraction of the recommendations of the PCAST reports, embodying them into administration policy.

00:59:27.160 - 00:59:40.920

The recommendations total, I can't believe whether it was Ashley or Jen, or Ashley and Jen in combination, who counted them up, but over 440 recommendations in those 39 reports.

00:59:44.760 - 00:59:57.000

Again, I'm not sure why we stress the length, but the longest ones... Chris, we had to write every page. Yeah, I guess it's sort of both a celebration

00:59:57.400 - 01:00:13.080

and a recollection of pain past that the 2012 report to the president on realizing the full potential of government held spectrum to spur economic growth, 192 pages, the report to which we just approved an addendum,

01:00:14.280 - 01:00:29.040

the report on forensic science and criminal courts, ensuring scientific validity of feature comparison message, that was 174 pages, and I believe entailed the review of several thousand articles, studies,

01:00:29.040 - 01:00:41.760

and reports by others to make sure we had overlooked no insight available from the literature on that topic. The 2010 report to the president in Congress on designing a digital future,

01:00:41.760 - 01:00:58.440

federally funded research and development in networking information technology, 148 pages. The shortest report, eight pages, triumph of compression, and Rosina and Michael and Maxine,

01:01:00.960 - 01:01:11.840

who led that study, deserve and are taking, I can see by their expressions and gestures, credit for producing... Hearing aids was pretty short.

01:01:11.840 - 01:01:25.120

The shortest report. I don't even know if it's short. Eight pages, but again, notable for its impact. Out of proportion to its length, in terms of the embrace,

01:01:28.600 - 01:01:44.240

not only by the president, but by the community affected of the recommendations on closer engagement of the federal government with the private sector on adaptation to the changes in climate that are ongoing

01:01:44.240 - 01:02:00.920

and will continue despite our best efforts at mitigation. This gets really hard to read, but this is just a listing in order of the reports produced by PCAST first term reports.

01:02:00.920 - 01:02:10.040

And then the second term reports, obviously I'm not going to redo the list, but it will again be up on the web. And those reports are all themselves posted

01:02:10.040 - 01:02:28.920

on the PCAST webpage in their entirety, along with the press releases that accompanied them at the time. So that's the quick statistical overview of what this group has been about for these eight years.

01:02:30.760 - 01:02:44.640

As Eric already noted in his opening remarks, a lot of thanks and credit are due, not only to the members of PCAST itself, but to the PCAST secretariat,

01:02:44.640 - 01:03:01.160

and to members of the OSTP staff. And indeed to scientists, technologists, innovators, across the federal government and across the society who have taken part, have fed insights into,

01:03:01.160 - 01:03:16.200

have participated in these PCAST studies. This has really been the work of a much larger body of people than just those sitting around this table. Indeed, more than 400 people served

01:03:16.240 - 01:03:24.520

as working group members alone, let alone all the other contributors. Let alone the other consultations and interactions that took place.

01:03:24.520 - 01:03:41.400

So in my view, this group has fulfilled its obligation to ensure that the science, technology, and innovation advice that goes to the president reflects the insights of the wider community,

01:03:41.400 - 01:03:57.760

and not just the folks who happen to be working full time at the moment in the federal government. So we would like now to expand a little bit on some of these achievements

01:03:57.760 - 01:04:13.320

and entertain as well comments from members of PCAST about the path forward, how we think about the role of science, technology, and innovation in government

01:04:13.360 - 01:04:23.800

after this particular administration and this particular PCAST have passed from the scene. So I think I'm gonna turn it over to Eric Lander to lead that particular discussion.

01:04:23.800 - 01:04:35.640

Okay. Because I know he's ever ready to do so. Well, there are lots and lots of different topics we worked on.

01:04:35.640 - 01:04:48.240

And I think it would be a fitting thing to close our meeting to just talk a little bit about the topics and their impacts. I note that the impacts we've sought to have

01:04:48.240 - 01:04:59.880

are sometimes within the federal government and sometimes outside the federal government. We're always providing advice to the president, but sometimes the advice to the president

01:04:59.880 - 01:05:11.960

we're providing are about things that federal government itself can undertake or things that the federal government can undertake in partnership with the private sector.

01:05:11.960 - 01:05:22.320

And in some cases we're providing insights about how one might think about the problem to the president, but those insights are intended

01:05:22.320 - 01:05:31.960

for a broader community as well. And very frequently it's others outside the federal government who are in the position to pick up and run with those insights.

01:05:31.960 - 01:05:46.000

And we've seen that happen again and again and again. So with no particularly planned order, but a little bit of a focus here, I thought I might turn to a few of our colleagues

01:05:46.000 - 01:06:02.120

and ask them to say a few words about some of these studies. I'm gonna pick on Mark Gorenberg first because the spectrum policy study that Peacast undertook and published in July of 2012

01:06:02.120 - 01:06:11.840

was a particularly impactful report and one I think that surprised us by its impact. When we started that study, there was a broad agreement amongst the group

01:06:11.840 - 01:06:23.800

that was doing it and the leadership of Peacast that it was unlikely to, it was likely to fall upon deaf ears that people didn't want this message.

01:06:23.800 - 01:06:34.080

And in fact, it turned out not to be the case at all. Mark? I think the timing of it was actually very, very good. I give Craig Monday a lot of credit

01:06:34.080 - 01:06:46.320

for coming up with the idea actually to pursue this for exactly that reason. This was really quite a team effort by both Peacast and the working group.

01:06:46.320 - 01:06:59.880

Peacast members, Craig, Eric Schmidt played a big role here as well as Jim Gates, Maxine and Bill Press in terms of our work. And really we called on everyone who all pitched in

01:06:59.880 - 01:07:09.480

and had a hand. And the reason why this was really timely is because spectrums have been allocated to the government and industry

01:07:09.480 - 01:07:21.120

in the same way now for 100 years, really since the sinking of the Titanic when we understood the importance of communication. And that's been this idea of finally dividing it

01:07:21.120 - 01:07:35.080

into exclusive frequency assignments and allocating to those particular assignments which is not very efficient. And also is very difficult once you're changing industries

01:07:36.680 - 01:07:46.120

as we've seen now from TV broadcasting to wireless, just the whole movement to that. It's extremely difficult. At the same time, the federal government

01:07:46.120 - 01:07:56.160

is doing their mission in many bands, probably well over 50% of the spectrum that's been allocated. And so spectrum now has reached a point

01:07:56.160 - 01:08:09.200

with this wireless revolution of not being available and not being reproducible. You just can't expand on it. So we looked at this 100 year change

01:08:09.200 - 01:08:22.920

of creating a new model, effectively this model of dynamically sharing it with the commercial and the federal sector. And that has a number of reasons of positiveness.

01:08:23.600 - 01:08:36.640

But one is that it can increase the efficiency and effectively capacity use of spectrum by thousands of times. And economically in what is being produced now,

01:08:36.640 - 01:08:47.680

that could lead to potentially worldwide trillions of dollars. So it had particularly a lot of impact. I think what surprised us was how quickly

01:08:47.680 - 01:09:00.800

both the US government and industry has responded to the report that came out in 2013, followed the President's memorandum in 2012 and the President's memorandum that came out in 2013,

01:09:00.800 - 01:09:13.480

which is that through tremendous work, the FCC commissioners approved the rule making, which was a painstaking process, finally in April of 2016,

01:09:13.480 - 01:09:25.760

but something that they voted unanimously to start just after our report in December, 2012. And that created a whole new concept of a band called the Innovation Band,

01:09:25.760 - 01:09:37.240

the Citizen Broadband Radio Service. And so that was 150 megahertz that had been put aside, that the government and industry had given up on basically for sharing.

01:09:38.560 - 01:09:49.120

And that's now in its way. The DOD has been a leading force in its opinion to move towards sharing. They created the National Spectrum Sordium,

01:09:49.120 - 01:10:00.600

which now has 115 industry members and a lot of funding to look at new technologies. And the White House has really embraced this launch of the Advanced Wireless Research Initiative,

01:10:00.600 - 01:10:16.720

which is really being led today by the NSF. But industry has been a huge positive sway in moving from the early incumbents of spectrum around voice to really this notion of data and IoT

01:10:16.720 - 01:10:28.680

and whole new sets of services that are important for wireless. And the core of our report was the creation of what we called a spectrum access system, which was the idea of a three-tier model

01:10:28.680 - 01:10:40.520

of allowing the government to have an incumbency, allowing people to have licensed spectrum in the middle layer and allowing a bottom layer to be general access. So it had the best of both worlds,

01:10:40.640 - 01:10:53.640

both quality of service at the license level and innovation at the general access level following on the innovation we've seen on licensed spectrum. Seven vendors in industry now have come forward

01:10:53.640 - 01:11:05.560

to build spectrum access systems just in this short period of time, and both Alphabet and Federated Wireless are ready to go. They've also even already done interoperability studies

01:11:05.560 - 01:11:15.040

between spectrum access systems, so they know this will all work. Chips are being developed for this already by companies like Qualcomm and the 3.5 band.

01:11:16.040 - 01:11:25.400

The Wireless Innovation Forum, which has 100 industry members, was the first endorser of the report, now has 277 participants working on this concept

01:11:25.400 - 01:11:37.000

and have regular calls, and 200 engineers were on their last regular call. So industry really is responding in the work that they're doing to make protocols and procedures

01:11:37.000 - 01:11:47.040

for device operation real. A new group was formed called the CBRS, CBRS being the name of the band Alliance, 24 members now,

01:11:47.040 - 01:11:56.280

and they're gonna be performing joint trials in the next 18 months. And then the third piece of this is how this is affecting the rest of the world,

01:11:56.280 - 01:12:04.560

how this US leadership is affecting the rest of the world, which is OFCOM, which is the FCC equivalent in the UK, has called for comments now,

01:12:04.560 - 01:12:13.440

they implement this in UK spectrum as well, 400 megahertz available, and they're going through that process today. And China, China Mobile,

01:12:13.440 - 01:12:22.400

is working with the US FCC to be able to harmonize the band. And the beauty of that is that it will open up another huge market.

01:12:22.400 - 01:12:32.440

So people who are developing chips and people who are developing devices will have a huge market to build into, and that will really expand the industry.

01:12:33.440 - 01:12:44.880

I think the most important point that I wanted to make here is that the report really worked towards the idea that spectrum was really scarce, but that doesn't mean that the spectrum itself

01:12:44.880 - 01:12:56.160

should become the investment, the value. What we wanted to do was come up with a way that made spectrum effectively so abundant that everything on top of it created the value

01:12:56.160 - 01:13:04.960

and created a whole new industry, again, worth trillions of dollars, as opposed to potentially the money that the government could make by reselling spectrum.

01:13:06.280 - 01:13:16.440

And we believe that this has the impetus to really change the 21st century for communications of all kinds. So with that, I'll turn that over.

01:13:16.440 - 01:13:26.760

Fantastic. And thank you for your tremendous efforts in this area. So I'm gonna suggest, I'm gonna touch on hearing aids.

01:13:26.760 - 01:13:35.400

Chris Castle should do this by all rights, because she was driving force behind it, but she's unable to be here this morning due to a personal obligation.

01:13:35.400 - 01:13:47.600

And then just to give a little advanced warning, I'm gonna ask Maxine to touch on advanced manufacturing. I'm gonna turn, if it's okay, to Susan Graham to say something about big data and privacy

01:13:47.600 - 01:13:58.760

and about the knighted reports that we've done. And then I'll give advanced warning to a few more people as we guys, I'm gonna turn back to Mark Gorenberg to talk about cities.

01:13:58.760 - 01:14:12.680

So that's a little advanced warning and I'll keep giving a little advanced warning since we haven't orchestrated all of this in advance. Hearing aids was perhaps one of the

01:14:12.680 - 01:14:26.640

most impactful reports we've done. And I think the most impactful report per page that we've done, it was an extremely short report. And it provoked a tremendous amount of action,

01:14:26.640 - 01:14:35.200

which we've seen just in the past couple of months. Problems are very straightforward. 30 million Americans need hearing aids. They need assistance.

01:14:35.200 - 01:14:46.360

If they do not get those hearing aids, they're really serious medical consequences, social isolation, falls, cognitive decline. And yet, more than two thirds of the people

01:14:46.360 - 01:14:53.400

who need them don't get them. Why? Because they're extraordinarily expensive and difficult to get.

01:14:53.400 - 01:15:04.920

These devices, if you need one in both ears, as many people do, can run you $4,000 to $5,000. And it's not covered by Medicare. It's not covered by Medicare

01:15:04.920 - 01:15:12.680

because they cost you $4,000 to $5,000. And if you multiply that by 30 million Americans, it's not affordable. But when you take a hearing aid out of the ear

01:15:12.680 - 01:15:22.000

and you look at it, it's a piece of plastic with not an awful lot of technology in it. And you look at all the amazing advances that happen in consumer electronics today

01:15:22.000 - 01:15:31.920

and you say, why isn't this thing that costs 10 times more than my iPhone at least as smart as my iPhone? There's a big market for these things.

01:15:31.920 - 01:15:45.040

Why isn't it something where I can put it my ear and I can tune it with software on my phone, for example, and optimize it and have feedback there? Why aren't there all sorts of innovative companies

01:15:45.040 - 01:15:56.920

jumping into this market to serve Americans and others around the world with really advanced devices that are smart, listening to the noise that's out there, filtering it in smart ways,

01:15:56.920 - 01:16:08.040

tuning it to what your own personal hearing needs are? Well, it turned out the reasons are that we have a set of constraints, a set of regulations that were put in place

01:16:08.040 - 01:16:19.200

in a different era when they made a lot of sense, but they don't make a lot of sense anymore. These devices, which should be like consumer electronics devices, at least in the case of basic hearing needs,

01:16:19.200 - 01:16:32.320

we're not talking about severe medical problems and plantable devices. We're talking about the thing you'd put in your ear externally, these audio conduction devices, air conduction devices.

01:16:35.280 - 01:16:46.960

These things should be innovating very much faster, but they're regarded as a prescription device. You've got to get a prescription from an audiologist or sign a waiver that you haven't gotten a prescription.

01:16:46.960 - 01:17:01.400

They have to be approved by the FDA in particular ways. They have to satisfy good manufacturing procedures. This shouldn't be. So we wrote a short and simple report that said,

01:17:02.560 - 01:17:13.960

for basic, mild to moderate, age-related, bilateral hearing loss, these should be over-the-counter devices that do not require a physician's prescription.

01:17:14.960 - 01:17:27.960

The National Academy of Sciences, NRC, conducted a study that came out several months after ours that was very similar to its conclusions. And to its enormous credit, the FDA has heard this loud and clear.

01:17:27.960 - 01:17:40.960

They've already taken the action of essentially abolishing the need for an audiologist's test. The way the FDA does that is the FDA says it's going to exercise its regulatory discretion

01:17:41.960 - 01:17:50.960

not to enforce the requirement. But that basically means you don't have to go to an audiologist, start now, and they're beginning the process to make this an over-the-counter device.

01:17:52.960 - 01:18:09.960

It's a fabulous example of a case where the federal government is able to step in, recognize that the industry has changed, that we need to open it up to lots of innovation, and to do so in a remarkably rapid and thoughtful way.

01:18:09.960 - 01:18:24.960

So I think we all feel really good about it, not least because, selfishly, but not least, because all members of PCAST are at risk for eating hearing aids at some point in the future, because as one age is,

01:18:24.960 - 01:18:37.960

almost everybody's going to need them. So in this one case, we're going to reap a benefit, but so will so many people around us, our parents right now. So that was the effort, and it all arose from a broader project

01:18:37.960 - 01:18:57.960

that PCAST was engaged in involving technology and aging. Chris Castle, who's our PCAST member, who's a gerontologist, had pushed the really great importance of focusing on technology to help aging Americans, and this larger report

01:18:57.960 - 01:19:11.960

from which this nugget was produced rapidly as a letter report, this larger report covers a number of other subjects, including just to pick one nugget out of that one, banking, the idea that as people get older,

01:19:11.960 - 01:19:25.960

there's a great deal of risk of fraud in their bank accounts. If you're using your credit card and there are funny transactions on your credit card, there are all sorts of software that can say, that was a funny transaction.

01:19:25.960 - 01:19:35.960

I'm going to hold it up and make sure that you actually meant to make that transaction. If somebody was draining your bank account, an elderly person who wasn't on top of all that,

01:19:35.960 - 01:19:48.960

and someone was draining your bank account with a series of unusual transactions, there are not either the software algorithms in place to monitor that, nor the general expectation in the industry

01:19:48.960 - 01:20:00.960

that banks should be looking after that on behalf of their customers. Well, because of that, as just one of the many parts of the aging report, that's begun to shift. The banking industry has embraced the fact

01:20:00.960 - 01:20:16.960

that that is a customer service that is technologically quite feasible. You can recognize, at least in many cases, bogus, suspicious transactions. You can delay them to make sure that they really were authorized. And I think the banking industry increasingly says,

01:20:16.960 - 01:20:30.960

we want to provide that service to customers. So that is happening. In this case, it's not clear that government regulation is going to be needed. It's the government stimulation that's going to be needed,

01:20:30.960 - 01:20:41.960

although it may come to a point at some point that, in fact, there are rules adopted that people should do it, but we're seeing much faster market adoption. There are many other things in this report on aging as well,

01:20:41.960 - 01:20:54.960

and I think we should be very proud of that. I'd like to turn to Maxine on advanced manufacturing, and then I've got Susan and Mark, and then I'm going to turn to Chris Chiba to talk to us about microbiological threats.

01:20:54.960 - 01:21:16.960

Thank you, Eric. As everybody knows, in 2009, we were recovering from major recession. And in 2011, PCAS decided that there was a real need for a concerted whole government effort to increase the footprint of applied research and development

01:21:16.960 - 01:21:36.960

in advanced manufacturing and cross the gap for how do you scale these technologies that are often beyond the risk limit of large manufacturing firms and particularly the smaller manufacturing firms. So there was a study in 2011 co-chaired by Eric Schmidt,

01:21:36.960 - 01:21:52.960

who unfortunately couldn't be here today, and Shirley Ann Jackson, who is president of RPI, who had previously been on PCAS, and there have been two subsequent studies in 2012 and 2014

01:21:52.960 - 01:22:18.960

that have been co-chaired by Alex Loveris, the CEO of DAO, and the president of MIT, first Susan Hockfield and now Raphael. And it's really a joint, the idea was to have a joint academic industry partnerships, and so we decided to create the National Network for Manufacturing Innovation Program.

01:22:18.960 - 01:22:37.960

It's now called Manufacturing USA, and it's also decided that this would be tried out by three major agencies, Department of Defense, which really needs the newest of technologies, Department of Energy, and the Department of Commerce, which has its role,

01:22:37.960 - 01:23:01.960

and NIST as part of the Department of Commerce is coordinating role, but also major coordination by OSTP and the Economic Council. So it's really been a very collaborative government program, and so in the four years that the program has been in existence,

01:23:01.960 - 01:23:25.960

which is the establishment of one institute in Youngstown, Ohio, by the Department of Defense, there are now 13 institutes. One was just announced this past week, and the federal government has committed for these five-year programs $920 million,

01:23:25.960 - 01:23:48.960

and the private sector and state and local governments have committed $190 million. And the intent is for five years, if there are to continue, there will be purely all non-federal government funding. And in addition, in 2015, there's a bipartisan action by Congress

01:23:48.960 - 01:24:11.960

against the revitalized American manufacturing and innovation to formally authorize the program. And so they're very exciting activities. So one, they're chaired by a non-profit, and something to be from like a smart manufacturing institute or other consortium group,

01:24:11.960 - 01:24:31.960

and then they're active participation by many different universities and also large spectrum of industries, both small and large, and city and local governments, and they provide this new one getting started that is sort of managed by UCLA for smart manufacturing,

01:24:31.960 - 01:24:53.960

and there's a smart manufacturing alliance that is sort of the lead of the non-profits, but you have UCLA and the city of Los Angeles providing a facility in the downtown area where actually there can be testing of some of the scale-ups, testing of non-destructive evaluation, all sorts of parts that will go into manufacturing

01:24:53.960 - 01:25:15.960

and through the whole value creation of manufacturing, not just the end product, but also supply chain and then regional areas. And in that one, there's five regional areas that, again, will consist of universities and industry-led RPIs, one of them Pacific Northwest Lab,

01:25:15.960 - 01:25:43.960

so you get various participation by also the National Lab, and these have already, groups have started to attract new manufacturing investment in the region. The oldest one in Youngstown, Ohio, which is looking at 3D printing, has attracted $32 million from GE for a printing hub,

01:25:43.960 - 01:26:15.960

and Alcoa has invested $60 million in New Kensington, Pennsylvania to do 3D printing with metal powders, and so it's really making, taking advantage of its mission and really pushing new technologies and new manufacturing, not just trying to revitalize the traditional of the manufacturing institute.

01:26:15.960 - 01:26:44.960

So we look forward, there will be coming out in the next month an evaluation of some of these programs, but just talking to some of the people, the enthusiasm and the real spin-offs are very, it's a very successful concept, and now all these players can work together, successful. That's great, thank you, and it's just amazing to see these 13 hubs that are just buzzing right now.

01:26:44.960 - 01:27:07.960

I think there's a recognition that, when we started, there was a great concern that manufacturing in the United States was perhaps lost to us, and I think what came out of that PCAST report was a recognition that low-end manufacturing, where the cost of labor was the major part of the goods,

01:27:07.960 - 01:27:26.960

was going to be a tough thing for America to compete on, because there were low-wage countries that would be able to manufacture things cheaply, but boy, there's just such an explosion we've seen in high-end advanced manufacturing that can't move elsewhere, nor frankly, where the wages are not the largest component of the goods,

01:27:26.960 - 01:27:49.960

and therefore there's not a downward pressure on wages in the way that there are in straightforward manufacturing. So these are good jobs at good wages, the sort of things we've wanted to have in manufacturing. We noted early in that report that Germany had gotten really good at doing that, and there was no particular reason why the U.S., sorry?

01:27:49.960 - 01:28:05.960

The Fraunhofer Institutes. But there's no particular reason why the United States shouldn't be the world leader in that kind of advanced manufacturing, and one could imagine that that will lead to a very large number of jobs, whether that will ever replace the whole number of jobs we had in the 1950s, who knows.

01:28:05.960 - 01:28:27.960

But I suspect it's already the case that there's a lot more of that that is going to be appearing in this country than we even imagined when we were writing the report, so thank you to everybody who worked on that. According to my list, we were going to turn to Susan to talk about, in whichever order she would want, the big data and privacy report from a technical perspective,

01:28:27.960 - 01:28:45.960

and the multiple reports that PICEST has done on the topic we call NIDERD, or Networking in Information Technology, R&D. So I'd like to start with NIDERD. The NIDERD Program, Networking in Information Technology Research and Development Program,

01:28:45.960 - 01:29:23.960

was started in the early 1990s during the Clinton administration, and it was created by congressional legislation to coordinate investments in information technology over the various government agencies that had an interest in IT. The legislation said that there was to be a biannual review of the program,

01:29:23.960 - 01:29:54.960

and there was to be an advisory function from the non-government community. About ten years later, that responsibility was assigned to PICEST. So in the beginning of this PICEST, a working group was created to do the review of the NIDERD Program, was co-chaired by David Shaw, who was then a member of PICEST,

01:29:54.960 - 01:30:22.960

and Ed Lozowska, who was at that time the chair of the Computing Community Consortium. And I participated in that review. So the review had two components. One was to review the coordination. The other part was to review the investment, what the R&D investment of the federal government was in Networking in Information Technology.

01:30:22.960 - 01:30:59.960

And in the investment part, we laid out emerging areas that we thought the government should be paying due attention to. One of those areas was privacy, and we said that privacy is not the same as cybersecurity. As more and more information is digitized, it's going to be an important issue for society, and the research community should be paying some attention.

01:30:59.960 - 01:31:24.960

We then did an update report in 2013. Peter Lee and I chaired that, and we looked at the recommendations from the previous report, and we said, well, some of these are getting appropriate attention, some of them are not. Gee, privacy is one of the areas in which you didn't listen to us.

01:31:24.960 - 01:31:56.960

We then did a review that came out in early 2015, which again looked at the coordination process, looked at the areas of investment and the emerging important areas of research, and in addition looked at the way that we understood what the investments were, which had to do with the way the reporting was done by the various agencies.

01:31:56.960 - 01:32:31.960

And we recommended some modernization of the reporting structure, and we got a very good response. Particularly in this last report, things changed in response to our recommendations, which was very satisfying. So that's the knighted situation. The coordination continues. Over the years, the agencies have learned to work together rather well,

01:32:31.960 - 01:33:09.960

and I would regard that program as a success story, which we have continued to monitor. I also might point out that the legislation says that the review should occur every two years. That's a little bit often, and that we now expect it's going to shift every three years, which makes more sense. So it was frustrating to some of us that we had twice recommended that there be more research in privacy,

01:33:09.960 - 01:33:36.960

and it didn't seem to be happening. And then in January of 2014, well, in the fall of 2013, the whole Snowden affair erupted. And in January of 2014, the president gave a speech in which he said, we need to pay some attention to privacy.

01:33:36.960 - 01:34:07.960

We need to find some ways of addressing this threat. And he asked for two reports. He asked PCAST to report on the science and technology of privacy and privacy protection, and he asked other people in the White House to report on the policy implications of threats to privacy

01:34:07.960 - 01:34:26.960

and what policy might be put in place. And he told us our role was not to talk about the policy issues, but to talk about the science and technology of privacy protection. And he wanted a response in 90 days.

01:34:26.960 - 01:34:33.960

So we put together a group. Craig was part of that group. Bill was part of that group. I apologize.

01:34:33.960 - 01:35:03.960

I don't remember who else among our PCAST members were parts of the group. And we set out to learn everything we could about privacy and to explain it in a report, which we actually did in 90 days. Part of that report was really a tutorial to the larger community about how the threat to privacy had emerged

01:35:03.960 - 01:35:28.960

because of information technology primarily. That more and more personal information is in digital form. People volunteer personal information because there's an implicit contract that if they tell this website something about themselves,

01:35:28.960 - 01:35:51.960

they'll get services in response that are services that they want. And some of them volunteer that information without really thinking about what they're giving away and to whom they're giving it away. And in addition, there's an increasingly powerful way to combine information from different sources.

01:35:51.960 - 01:36:18.960

And so you think you're only telling somebody what your gender is and what your age is, and yet that can be combined by your zip code and lots of other information. And we have increasingly powerful ways of analyzing data. So in fact, as a society, we can learn an enormous amount about people.

01:36:18.960 - 01:36:47.960

And we made a bunch of recommendations. We had a fair bit of exchange with the other group that was making policy recommendations. One of the key observations that our report made, and this was an outgrowth of a paper that Craig had written, was that you can't really prevent the collection of this data.

01:36:47.960 - 01:37:13.960

As I said, there are good reasons why the government collects some data, and there are good reasons why people volunteer data in return for services. And in some sense, every time you volunteer information, you're doing a risk assessment in which most people decide that it's worth the risk to get the services.

01:37:13.960 - 01:37:45.960

So the data's going to be out there, and the analysis is going to be out there. What you can control is the way personal information is used. In principle, you can control the way it's used. In order to do that, you have to again come back to using IT to help to control the use

01:37:45.960 - 01:38:18.960

together with policy that says certain uses are not appropriate and are not legal. And in doing that, you need to take into account the fact that different people have different views of what they regard as private information about themselves and their families. You need some way to be able to have different degrees of control depending on people's different value systems.

01:38:18.960 - 01:38:50.960

And there are some parts of that problem that we as a science and technology community know how to solve now. And there are some parts of that problem which we again throw back to the research community to say, figure out how to institute these controls in a way that we can manage them, in a way that they're affordable, in a way that we can check whether they're actually being applied or not.

01:38:50.960 - 01:39:25.960

So because we didn't make policy recommendations, we can't say that there was legislation in response to our recommendations, and yet there's been an enormous increase in consciousness, increase in knowledge, and increase in the ways that the commercial sector is able to grapple with the policy issues. I don't know if you could say just a little more about the way that these ideas are rippling out in the community

01:39:25.960 - 01:39:46.960

and in companies and things. Well there are now courses that use this report as part of their text material, and in that sense we're educating the next generation, we're educating lawyers, we're educating people who work on policy, and we're educating the technical community.

01:39:46.960 - 01:40:12.960

What was the other half of your question? Yeah, that's fundamentally it. The framing of the report that I think Craig distilled into, it's not productive to regulate collection, regulate use, more control use, was I think really important framing because there was, and still is in some places,

01:40:12.960 - 01:40:38.960

a notion that placing the controls at the point of collection would even be feasible, and it's just not, because you can't tell at the point of collection what all these pieces are going to be good for. They may look all, you know, minor. Well the other part of it that I should have pointed out is that a lot of the collection is not visible to us or under our control.

01:40:38.960 - 01:41:02.960

For example, if you think about the pervasiveness of cameras in our environment, we're photographed all the time, the face recognition is getting better and better, and yet most of the time we're not consciously aware that we're being photographed, and even if we were, there's no issue of consent.

01:41:02.960 - 01:41:27.960

If you're in the public, it's perfectly legal and okay for you to be photographed. Yeah, it's a good example of I think trying to reframe these issues in a way that both policymakers and technologists can work with them. Thank you. Let me turn to Mark on cities again, and then I'm going to turn to Chris Chiba on microbiology,

01:41:27.960 - 01:41:46.960

and Bill Press on cybersecurity. Eric, thank you. So again, the city's report came out of the brainchild actually, you know, of Craig Monday, and I hope that Craig will add some comments after I'm done with these sort of introductory comments.

01:41:46.960 - 01:42:10.960

In terms of studying what's happening in cities and how they're transforming and how the federal government can be helpful in that regard by adding policy to the technology. So Craig and Eric Schmidt were co-chairs on this report. Huge credit to Marjorie Blumenthal, who sat on top of this for frankly a year working on it with us.

01:42:10.960 - 01:42:34.960

But this is a report that I'd actually view as co-chaired by almost all PCAST. We did our first meeting publicly together in February of 2015 to pick a working group out in Palo Alto, and clearly everybody was involved, and everybody stayed involved. People like Jim Gates were, I think, on every single call for about a year.

01:42:34.960 - 01:43:02.960

Susan did a final review through the way, and everybody added their part because this is just so interdisciplinary in its work. And the interdisciplinary tone of this set the stage, or some of our recommendations, the need to break down the silos of government to become much more interagency, the need to create competitions, the need to look at cities through the lens of a district,

01:43:02.960 - 01:43:31.960

rather to be able to have something that is large enough to have meaning but small enough to experiment on. The ability to look at this in terms of districts improving on the dimensions of being more green, being more mobile, being more connected, effectively solving issues such as poverty. And so we also looked at the necessity of this to share information,

01:43:31.960 - 01:43:56.960

and came up with a concept in the report of the city web that is how to share information and insights across cities, because some of the larger cities are actually working on these problems today, but the smaller ones just don't have the budgets to do that and can learn tremendously from each other. The timing of this that really is the reason why it has potential for great impact

01:43:56.960 - 01:44:15.960

is that we're really entering into a fourth era of cities. I mean, we went from the era of the steam engine to the electric grid to then the personal automobile, and that's been true now since really before World War II, but particularly after World War II with the creation of the highway system.

01:44:15.960 - 01:44:41.960

And that's the way funding has been put in place, but now we're seeing this change, and that's because urban planning is now transforming, frankly, into urban science. It's because there's a demand that is appealing to 80% of our population live in cities, 90% of our country's GDP, centers around cities,

01:44:41.960 - 01:45:02.960

and this change happened around 2011 with the start of this sort of sharing economy with millennials, frankly, and baby boomers coming back to the city, and changing their norms of behavior, rallying around the shared economy, rallying around the concepts of access instead of ownership, rallying around the idea of living in less space,

01:45:02.960 - 01:45:28.960

and rallying around the idea of more walkability and bikeability and sort of less transit movement. And at the same time, we saw cities starting to get instrumented in terms of data and sensors, and we saw technologies all transforming together, like transportation. I mean, everybody knows about autonomous vehicles that are coming,

01:45:28.960 - 01:45:44.960

and it's actually real within the next five years, new energy systems, new building and housing systems, new district water systems. Maxine talked about in the manufacturing report new concepts around urban manufacturing, and we even see urban farming.

01:45:44.960 - 01:46:09.960

And so when you put all this together, you have the ability to really transform cities and to make a point of this in a way that's thinkable. About 25% of our downtowns now are used up for parking and tied to our old transportation system. And if you put in autonomous vehicles and you can get that 25% back,

01:46:09.960 - 01:46:26.960

frankly, 80% in some of the districts of downtown Los Angeles, you can transform housing into very different district housing. Once you create new housing around that, you can create new district energy systems. You can create new district water systems.

01:46:26.960 - 01:46:42.960

You can really transform the whole makeup of the districts of these cities, which has huge opportunity for revenue growth and huge opportunity for cost savings and huge opportunity for competition. And we're actually falling behind.

01:46:42.960 - 01:47:06.960

The Asian Import Bank is spending $18 billion a year on this. The UK has created essentially a cabinet member, a minister of cities. And cities like Stockholm and Hamburg and Amsterdam are transforming in public-private partnerships to a whole new model that's far ahead of us.

01:47:06.960 - 01:47:26.960

And that is both an opportunity, a chance for us to fall behind, but also an opportunity to work together. And so the report came out in February of 2016. And we had a lot of conversations with both industry and with agencies along the way.

01:47:26.960 - 01:47:43.960

And so a lot of this work started happening towards the end of our report cycle, but also since that report cycle. One of our key recommendations was around implementing competitions. The Department of Transportation has really taken a tremendous lead here

01:47:43.960 - 01:47:57.960

with their Smart Cities Challenge. They ended up getting 78 city proposals, far more than they expected. Seven finalists came out of that. And all of them, by the way, are working on their proposals,

01:47:57.960 - 01:48:15.960

although the winner, Columbus, Ohio, was the one that was funded by the federal government with $40 million from the Department of Transportation, which, by the way, then leveraged into another $100 million of private funding alongside of that. They chose four districts for change with the implementation of autonomous vehicles

01:48:15.960 - 01:48:43.960

and battery research, electric charging stations, buses to cars, new types of shuttles. But what was interesting about it was that they're also trying to solve the issue of infant mortality. The infant mortality level in Columbus, Ohio, is four times the national average. And that's because of the lack of accessibility of doctors and transportation into many parts of the city.

01:48:43.960 - 01:49:06.960

So part of these competitions, again, was to look at things like making cities more green around electric vehicles and electric charging stations, and the DOE worked with DOT on this, but also this idea of changing and fixing some of the poverty and connected issues of some make our cities more inclusive.

01:49:06.960 - 01:49:23.960

What I find most interesting about this competition is that it got so many people involved, although only Columbus was the one that was funded. You know, 1,400 people participated in these DOT webinars, and 800 people participated in their forums,

01:49:23.960 - 01:49:43.960

and 300 companies, industry companies, joined in their proposals. And so they and the other six finalists are all working on that today because of this small amount of money that had huge leverage. And in fact, DOT has kind of doubled down on this now because of its success.

01:49:43.960 - 01:50:01.960

We're at the Frontiers Conference in this past October. They gave out another 65 million in grants. That led to another 100 million in private sector funding. And they've now committed as part of the FAST Act, 240 million over the next four years in this area.

01:50:01.960 - 01:50:22.960

Beyond the Department of Transportation, NIST has moved forward with their Global Cities Team Challenge. They did, they working with U.S. Ignite hosted their supercluster, kickoff 150 cities and 400 companies around the world working on these issues. The White House, towards the end of our report as we work together,

01:50:22.960 - 01:50:45.960

the White House put together their first Smart Cities initiative in September of 2015, and they did their second round this year in 2016 with 80 million in funding from federal investments, led by new ideas in the NSF, 60 million from the NSF, 15 million from DOE, with the impact on 70 different city communities.

01:50:45.960 - 01:51:17.960

And they also grew their MetroLab network, which we really think is a model for how city CTOs and universities can partner to solve some of these issues around urban innovation. And now there are 51 universities and 38 cities working together in Paris to solve some of these horrible issues. We also called for coordination of research across the different agencies,

01:51:17.960 - 01:51:46.960

and that is well underway now, and we're expecting to see a report for public comment in this next week on how to put research together for urban science and urban innovation. And also there is a federal blueprint being worked on sort of cross agency right now. We're hoping to see that come forward, if not before the end of this administration,

01:51:46.960 - 01:52:19.960

by people who will remain to put that together as a blueprint that we're very hopeful the next administration will pick up the ball on. This is one of those studies and concepts that I particularly am very optimistic that the next administration will take charge of. The reason why is because, I mean, the president-elect has already talked about a trillion dollars in infrastructure improvement. We're hopeful that they'll look to some of the guidance here in how to spend some of that,

01:52:19.960 - 01:52:45.960

but we think this is a multi-trillion dollar business opportunity. We think it can create all kinds of new jobs and be actually one of those bright lights for creating jobs over the next few years. It can certainly enhance the quality of life for all its city residents, and we think it can improve infrastructure that's really critical for things like homeland security and resilience to climate change and disasters,

01:52:45.960 - 01:53:12.960

because that clearly is one of the tenants that can be implemented by these technologies that can be very impactful for our nation's city. So with that, I wanted to turn it over to Craig and anybody else who may want to make comments. Thanks, Mark. I mean, you were pretty comprehensive in covering many of the points, but I'll just emphasize a couple. One, there's been a lot of activity under the banner of smart cities,

01:53:12.960 - 01:53:41.960

which is more or less applying technology to improve the efficiency of existing city structures. I think one of the things that we emphasized in this report, and which I think people are beginning to adapt to, is that the steady progress in more of these infrastructural technologies like energy and transportation in and of themselves cause cities to change, that their nature changes, what urban living means changes.

01:53:41.960 - 01:54:15.960

These changes don't happen very often, but we believe that they're about to happen again. And, you know, people oftentimes underestimate how quickly these transformational things happen. If you go back to the decade just before World War I, New York went from an all-horse environment to an all-car environment in about 12 years. And more recently, I mean, you know, my own experience visiting China in 1987 was an all-bicycle environment in Beijing,

01:54:15.960 - 01:54:41.960

you know, and less than 15 years later was essentially almost an all-car environment, and the nature of those places change. We see these kind of changes coming now. Similarly, many of the infrastructures that created the urban evolution, as we've known over 150 years in this country, were each, you know, individual technological advances, for example, in energy or transportation, the car.

01:54:41.960 - 01:55:12.960

But now we have all of them in sort of their earliest incarnations, but the cities have lived on those for over 100 years. So many of them are basically being depleted or are going to be up for replacement anyway, and therefore we wanted to signal to people that the investment that is going to have to be made needs to be made in anticipation of these new infrastructures and not just a replacement of the old one in situ.

01:55:12.960 - 01:55:43.960

Also, the fact that we understand them as a system now is a very different way of thinking. So the district approach and talking about energy heating and cooling as one integrated thing, not three different things, you know, are very, very important ideas. None of these were really emergent out of the smart cities kind of approach, and yet, you know, we see a lot of work, particularly in the governments, who contemplate huge capital investments required to deal with aging infrastructures.

01:55:43.960 - 01:56:16.960

You know, the water problem that we also wrote a report on is essentially a byproduct in many ways of degrading classic infrastructure. And so I think we wanted to really get people to focus, not just on tweaking cities at the margin, but recognize that fundamental change is coming, is necessary, and we need to be prepared for that. And so I think that that was a big delta from classic thinking, you know, that came from this report.

01:56:16.960 - 01:56:56.960

Let's turn to Chris Chai about to talk about a constellation of reports that really emerge from various kinds of microbiological threats, including the very first report the President asked us to do. Thank you, Eric. It's true, there's been a kind of continuity through the eight years of PCAST with respect to attention to microbiological threats. Going back to our very first report that Eric mentioned on influenza in 2009 and extending up to one of our final reports on action needed to protect against biological attack.

01:56:56.960 - 01:57:38.960

After our initial report on influenza in 2009, a year later, we expanded, we provided much more context for what the nation had faced with the H1N1 influenza pandemic in 2009 with the 2010 report on re-engineering the vaccine production enterprise to meet the challenges of pandemic influenza. You know, the issue was that the time scale between the declaration of an urgent need for a vaccine and the availability of the first limited doses of that vaccine had been 26 weeks, which was eight weeks after the start of the second wave of the pandemic in the United States.

01:57:38.960 - 01:58:24.960

The PCAST 2010 report set as a goal and identified a suite of individual measures that could be taken towards that goal, set as a goal to reduce the timeline for vaccine production to about 12 weeks. There were a succession of other PCAST reports that were all in, that all had either the theme or had a component of the theme of preparing for microbiological threats. In 2012, there was the PCAST report on propelling innovation and drug discovery, which included, which examined issues of increasing the pace of development of therapeutic drugs, including the development of medical countermeasures against infectious diseases.

01:58:24.960 - 01:58:56.960

And then in 2014, PCAST released its report in combating antibiotic resistance. That report identified a suite of measures intended to increase the pace of development of new antibiotics, and it also measures intended to protect the utility of existing antibiotics, that is, by slowing the rate of growth of antibiotic resistance. The report endorsed a one-health approach, which says that you have to examine the situation in animals as well as human beings.

01:58:56.960 - 01:59:28.960

You have to conduct surveillance in both the human population and the animal population, and this has to be part of a comprehensive strategy. And then finally, most recently, we had their letter report to the president, which was one of our two reports that include a classified annex on protecting the United States against biological attack. I co-chaired that report with Wanda Austin. I think the continuity through eight years of PCAST on confronting microbiological threats shows the ongoing importance of these issues.

01:59:28.960 - 01:59:52.960

It's clear looking forward that the importance of these issues is never going to go away. This is something our civilization has to face. Our country has to face, and our civilization has to face for the foreseeable future. In fact, I think it's likely that these challenges, both with respect to natural disease and with respect to potential human-made threats, are only going to increase in the coming years.

01:59:52.960 - 02:00:32.960

Let me summarize the unclassified parts of our final report on actually needing to protect against biological attack. As has been often the case, almost always the case with PCAST reports, we brought together a working group of specialists, microbiologists, and virologists. We also had strong PCAST participation in addition to the co-chairs, Bill Press, Barbara Shaw, Eric Lander, John Holdren. The basic idea of the report, and I think the basic message, is that advances in technology are now so important, both quantitatively.

02:00:33.960 - 02:01:04.960

If you define, as your metric, the time it takes to synthesize an oligonucleotide, a short strand of DNA, you see an increase, an exponential increase in biotechnological power. That's comparable to the computer power increase explained by Moore's law, or described by Moore's law. And that's the old Moore's law, not the new slower rate of Moore's law. But now you also see qualitative jumps with new DNA technologies coming online.

02:01:04.960 - 02:01:45.960

And you put those trends together, and the clear result is that the approaches we have taken towards biological security, and in particular biodefense, are no longer going to be entirely sufficient. It's no longer going to be sufficient to focus on a list of select agents and possible manipulations of those agents. There's going to be a potentially much broader landscape that an intelligent adversary could choose to engage on with respect to the creation of non-biological threats. Government is always inherently challenged to keep up with rapidly moving areas of technology.

02:01:45.960 - 02:02:15.960

That's because the people that are developing those technologies are primarily in the private sector and academia. And therefore, we need to have mechanisms to constantly inform the government. And this has to be true in perpetuity, to make sure the government remains on top of these new technical developments. So our report divided our recommendations into immediate, medium term and long term steps.

02:02:15.960 - 02:03:00.960

And all of those steps, though, need to begin now, because the things we want to have in place for the long term, let's say 15 years from now, need to begin now if they are in fact going to be in place 15 years down the road. Absolutely crucial is the issue of leadership. White House leadership, above all, because this problem cuts across so many agencies, leadership within a variety of agencies, including most important, perhaps, the intelligence community. That leadership is going to have to establish, is going to have to build on the already existing strategy for biodefense to ensure that it is routinely, annually updated,

02:03:00.960 - 02:03:36.960

taking into account new advances in the technology and the ramifications of those advances. And that leadership is going to have to hold to account the various agencies to make sure that they are in fact fulfilling their responsibilities with respect to that strategy. And there has to be a kind of back and forth routine communication between the White House leadership and the various agencies engaged on this issue. Another immediate step that we called for was the creation of a public health emergency response fund.

02:03:36.960 - 02:04:00.960

You know, Congress eventually appropriated $7.7 billion for the 2009 H1N1 influenza outbreak. It appropriated $5.5 billion to combat Ebola. There is no question that when there is a major outbreak in the United States or internationally with clear, with presenting a clear threat to the United States, that substantial funds are going to be appropriated.

02:04:00.960 - 02:04:34.960

What one wants the federal government to be able to do is to be able to move out of the gate quickly to get out in front as fast as possible. And that will, in fact, minimize the amount of spending required by the federal government, as well as, of course, minimize the impact on the health of Americans and health of people around the world. We have an analogy, an analog for this, which is the Federal Emergency Management Agency's Disaster Relief Fund. When there is a hurricane or a tornado, FEMA doesn't have to, the executive branch doesn't have to go to Congress, FEMA doesn't have to go to Congress,

02:04:34.960 - 02:04:59.960

and ask for a new appropriation to allow them to respond to that catastrophe. The money exists, already appropriated, and a fund that FEMA can immediately draw on. It has to report monthly to Congress as it spends that fund down, and as the fund is spent down, Congress replaces the funding. The important thing is that one can act immediately.

02:04:59.960 - 02:05:27.960

By analogy, we call for a public health emergency response fund that the executive would be able to draw on immediately in the event of a major pandemic, rather than having to wait for a particular crisis, and again, rather than having to wait for a particular appropriation. And again, we know historically those monies will be appropriated. You know, those are congressional research service numbers for the H1N1 and the Ebola outbreak.

02:05:27.960 - 02:05:53.960

Substantial funding will be appropriated. The amount of funding required will be less if we can move out faster. In the medium term, we need to continue to increase our ability to conduct biosurveillance, again, taking a one-health approach. So this has to happen both with respect to agriculture and with respect to human health. We note that the Centers for Disease Control domestically and internationally, both the CDC and the Departments of Defense,

02:05:53.960 - 02:06:20.960

are improving biosurveillance and are adopting genomics, genomic approach to surveillance. One wants to rapidly understand the genetic diversity of whatever organism one is facing, look at the various strains of the organism, and determine whether the organism has been modified, intentionally modified, and you want that to be able to happen as quickly as possible.

02:06:20.960 - 02:06:48.960

So biosurveillance needs to be a continuing area of emphasis, and the already underway movement of biosurveillance towards the ability to apply modern genomic techniques needs to be continued. We also, in both the medium and long term, set goals for improvements in medical countermeasures. Once you recognize that something is underway, whether it's a natural threat, a naturally occurring threat, or a human engineered threat,

02:06:48.960 - 02:07:11.960

we have to be able to respond quickly. And if we look down the road at what will be increasingly available to potential adversaries, whether we're talking about non-state actors or state actors with the technology and the ability to manipulate organisms, we had better be able to act increasingly rapidly with respect to medical countermeasures,

02:07:11.960 - 02:07:47.960

and therefore we set two ambitious goals, very ambitious goals, in the PCAST letter report to the President. One is that on a time scale of a decade, the United States needs to be in a position where it could have the ability to accomplish, in a six month period, the complete development, manufacture, clinical testing, and licensure of a vaccine against a new threat. Everyone recognizes that that is a very heavy lift, but yet in the world that we are entering, it's one that we need to work towards rapidly.

02:07:47.960 - 02:08:18.960

And then on a longer time scale, we need to be able to respond even more quickly with medical countermeasures, with at least therapeutics to outbreaks. Ultimately, that's something that we want to have the ability to have therapeutics available to the American population within days of the recognition of a new threat. That is a kind of blue sky requirement, but it is one that one can already begin to identify a handful of possible approaches to pursue,

02:08:18.960 - 02:08:56.960

and it's time to get underway with trying to make those a reality. So in conclusion, Eric, I'll just say that I'll reiterate that PCAST has recognized there's been a kind of steady drumbeat on a kind of biannual basis of this recognition of the microbial threat. There was a time decades ago when people talked about a post-infectious disease era that was in retrospect a kind of naive hope. Microbes in the natural world evolve, and we now know that even if microbes are eliminated from the natural world that human beings,

02:08:56.960 - 02:09:21.960

as a consequence of biotechnology, which has brought so many advances and gives us so many opportunities to protect ourselves, but those technologies have also brought with them the ability to recreate microbes that were once eliminated from the natural world. This is a challenge we will face forever. I'll conclude by saying I need to say something about, even though it's such early days with this report, something about reaction in the community,

02:09:21.960 - 02:09:44.960

I will note that in addition to great interest in the executive branch, there has been clear interest on Capitol Hill. We have briefed staff from half a dozen different congressional committees on this report. This is clearly a bipartisan issue, and one we all recognize will face through the future. Thanks so much, Chris. That's very, very helpful.

02:09:44.960 - 02:10:04.960

We'll take just a couple more of these because I think we're giving the sense that the goal is not to be comprehensive. I asked Bill to say something about the cybersecurity report. Rosina, if you would just take a couple of minutes on the water report, we'll keep them brief, and Jim Gates, if you would just say a couple of minutes about this whole series of education reports we've done

02:10:04.960 - 02:10:48.960

that might bring us to the point where I think we've covered enough to get a representative flavor of what has flowed through. Apart from then turning to our chair, our co-chair here, John, to talk, I thought would be fittingly last about climate and energy. So the subject is cybersecurity. There are some national problems that are so difficult and so fast-moving that even PCAST was not able to solve them in the eight-year period that we were engaged in this issue.

02:10:49.960 - 02:11:13.960

I say that a little bit jocularly, but only because if you open up your newspaper headlines today or yesterday or the day before, cybersecurity dominates those headlines. And I think it's important that we recall that this has all happened in the last eight years, or much of this has happened in the time that we've been on deck here.

02:11:13.960 - 02:11:46.960

If you can think back eight years ago, cybersecurity was something that the guys and gals in the IT department worried about a little bit, and mid-level management and corporations worried about a little bit. And for the average citizen, the kind of issue was, how do I stop all the spam I'm getting in my email, and, oh, yes, there's this idea that maybe my computer is being co-opted for somebody's bad use, but that's not very likely.

02:11:46.960 - 02:12:17.960

Now, fast-forward to today when people's identities are stolen, people's bank accounts are drained. This is an issue that has gone from being arcane and specialized to literally at the forefront of American government, and as we have seen to much regret, American politics. So where has PCAST been in all of this?

02:12:18.960 - 02:12:47.960

But I think there's also a lot of necessary context here, because it isn't purely a technical problem. The question of how we as a nation defend ourselves against cyber threat, against cyber attack, is something that stresses our whole notion of what is the dividing line between what government is responsible for and what the private sector is responsible for.

02:12:47.960 - 02:13:21.960

You know, if you think of cyber attack, particularly cyber attack from abroad as being an attack on us as a nation, then you would conclude that we ought to be defended down to the level of every individual citizen by our military. In fact, if you go back more than 100 years in American history, the military was used to defend within the borders of the United States. It was not a very happy period in American history.

02:13:21.960 - 02:14:11.960

The military was defending primarily white settlers of the West against indigenous populations whose land was being taken away. That's not the point. The point is the federal government accepted that the military should defend its citizens even within the borders of the United States. So the modern analog would be an agency like NSA or DOD that has extremely capable defensive cyber security capabilities. But we by national policy limit the use of those capabilities to defending a very small circle of intrinsically governmental things.

02:14:11.960 - 02:14:44.960

The NSA defends, roughly speaking, the dot mill network. By executive order, the government turns to the Department of Homeland Security to defend by and large the dot gov network. There's then a kind of a gray area much debated of what is critical infrastructure, which lies in the private sector, but whose disruption would be so fundamental to our economy, to our life that government does have a clear role.

02:14:44.960 - 02:15:23.960

So that starts being things like the financial sector, power, electricity. But even here, the government starts, in effect, turning over the responsibility for this defense to the private sector itself. And then when you finally get down to almost all of our economy, the government basically says to us, says by policy, you're the private sector. This is not really different from you should build your own fences or put your own locks on your doors and windows to defend your business.

02:15:23.960 - 02:16:03.960

The role of government is limited to prosecuting crime. It comes down to the FBI point of view. So I know from having participated in discussions with people from other countries that other countries don't view this as a rational division of labor to cybersecurity. They say, why aren't you using your best towns all the way down the value chain as it were, or all the way up the value chain? Interestingly, of course, this goes back in American history to the election of Rutherford B. Hayes in 1876.

02:16:03.960 - 02:16:37.960

I don't know if my colleagues know this. Because in part of the compromise between Republicans and Democrats, then quite different parties from today that put Rutherford B. Hayes into office, there was an agreement to withdraw all federal troops from the south. And there was a law, the Posse Comitatus Act passed that forbid for all time the use of federal troops of what we would today say is the military

02:16:37.960 - 02:17:23.960

in a whole range of civilian defense functions which have come down to us today as including cyber defense other than in a very narrow defense of government or presumably if there was a Pearl Harbor-like event, a catastrophic event of a magnitude in which the president could invoke special powers. So it's in that kind of a strange political context extending back a century that PCAST in the last eight years has tried to be helpful in its cybersecurity recommendations. And let me give you examples of some of them and try to assess whether we've done any good.

02:17:24.960 - 02:18:16.960

In our principal unclassified cybersecurity report, the overarching finding was that cybersecurity could not be achieved by any set of static precautions, whether taken by the government or taken by industry, static checklist approaches to cybersecurity were doomed to fail. Rather, a national approach to cybersecurity at all the levels I've described has to be a set of processes that continuously couple information about an evolving threat to defensive reactions and responses. And I would say that recommendation which we issued about the middle of our tenure in the last eight years certainly stands the test of time.

02:18:16.960 - 02:18:45.960

I think we all think that that's true today. I think that the government organizations that I've described, particularly Department of Homeland Security, are coming around to that view. Coming around to the view that you don't try to draw a line and say above this line is critical infrastructure and we're going to regulate critical infrastructure by giving them a checklist of mandated things they must do and below this line good luck.

02:18:46.960 - 02:19:33.960

Instead, what we want to do is say, and NIST as a government agency has been the most active in this, NIST has said, let's bring in industry people at these different levels, some in critical infrastructure, some in simply the economy at large, and let's get the individual industries discussing what are the appropriate level of cybersecurity measures that they can take. And much more importantly, as our overarching finding says, what is the process by which we will continue to evolve this framework so that as the threat evolves, the accepted standards for any given industry also evolve.

02:19:33.960 - 02:20:19.960

When we issued our cybersecurity report, one of the things we did, I know Craig and I remember this vividly, was go and visit the folks at the SEC, the Security Exchange Commission. Why the SEC? Because the SEC is almost the only federal agency that can ask, can require the whole economy or some very large part of it, that is to say all publicly held companies to do things. And we said to the SEC, here's our cybersecurity report, it's advice to the president, they said to us, go away kids, we don't take advice from the president, we're an independent regulatory agency.

02:20:19.960 - 02:21:04.960

I'm paraphrasing what they said, of course. And we said, we only want to be helpful, and we think that this threat is moving so fast that you will be drawn into this space, that your mission of protecting a level playing field for investors is going to draw you into the cybersecurity space. And they said, well, okay, leave a copy of your report and maybe we'll think about it. Over the next, I guess it's now been three to five years, the SEC and several other government agencies have moved into this space and have moved into it in a way that I think we find reassuring.

02:21:04.960 - 02:21:38.960

I think that maybe we weren't the right people to be advising the SEC, but the cascade of events is sufficiently compelling that cybersecurity is on the agenda in every boardroom of every publicly held company in America now. And with an approach which is by and large moving away from checklists and towards the kind of continuous improvement. That's great. So I realize our time is drawing short.

02:21:38.960 - 02:22:00.960

And so I'm going to actually ask that Rosina give the five sentence summary of the high points of water. Jim Gates, and it's the staff of life. You have three minutes. Jim Gates can get a five sentence version that I want to make sure John has a chance to comment on climate and energy.

02:22:00.960 - 02:22:23.960

Okay, I had quite a bit of opportunity at the last meeting to have you approve our recommendation. So let me just give you a very quick update. The full report and all its glory was delivered on December 19th, kind of a Christmas present to the administration, but there have been a number of positive things that have happened. I mentioned that we thought about short term as well as long term and let me just say some of the short term things that are happening.

02:22:23.960 - 02:22:58.960

One, EPA released its drinking water action plan, which is very synchronous with our recommendations. Two, Congress passed the water infrastructure improvement for the nation act, which allocates 200 million in low interest loans to upgrade infrastructure across the nation as well as 100 million to Flint. And some of our short term recommendations are actually being implemented, even as we speak or very soon. One was that we suggested EPA revisit the idea that if only fewer than 10% of your samples exceed 15 parts per billion, all was okay.

02:22:58.960 - 02:23:30.960

Of course, that means there can be some very high numbers. And as they are revising the lead and copper act, they are taking this under advisement and we expect that that will change. Citizen science was one of our recommendations and the number of agencies through their coordinators are already coming up with new ideas for citizen science and water. And finally, we discovered three multi-agency groups focused on water and they have come together and are coordinating existing data and working through two existing portals to make it more accessible to the public.

02:23:30.960 - 02:23:37.960

So we feel in only a few weeks we've already had some big successes. Fantastic. Jim Gates. Thank you.

02:23:37.960 - 02:24:02.960

And I will indeed be brief. First of all, there were a sequence of four reports that PCAST did on STEM education, first from being prepared and inspired, the next engaged to Excel. We did report on IT, MOOCs basically, and then a final report on jobs and skills. I think probably for me, the most of the high points that I would like the public to take away as well as our PCAST members is that the education,

02:24:02.960 - 02:24:44.960

every, sorry, every student succeeds act, which was passed in December of last year, was heavily impacted by our discussions of STEM. There are parts of that report, particularly section 4107, where you will find that ideas that were in our STEM and report have been passed. This represents, I think something that's rather interesting in this past few years, this is a bipartisan victory for PCAST recommendations, which I find just extraordinarily as an accomplishment. The other thing I would tell you is that I would point back to the involvement and feedback of teachers, because one of the things that was passed in there was support for a concept that came from teachers themselves.

02:24:44.960 - 02:25:29.960

This is the STEM Master Teacher Corps, this is a state led effort, and this was called out to PCAST by the teachers who had been awarded the President's Award for Excellence in Math and Science Teaching. So this is a closed loop where a group of citizens spoke to the government and impacted the law. So I will just end by reminding some of you that perhaps in your childhood you too saw the alphabet rock on a bill going to Congress and becoming a real accomplishment. We did it, folks. Very good. Very good. John, we're going to give you the last word on a topic that is very important and one that you've played such a central role in of energy and climate.

02:25:29.960 - 02:26:42.960

So on energy and climate, PCAST certainly played a very substantial role in helping to shape this administration's response to those interlocking challenges. The November 2010 PCAST report on energy technology innovation in the context of a comprehensive energy strategy led the Department of Energy to produce the first quadrennial technology review a year later in November 2011 and a second one appropriately four years after that in late 2015. These have been the most comprehensive and in-depth looks ever at the array of energy supply and energy end-use technologies, their status, the possibilities for accelerated innovation, what would be needed in the way of roadmaps to move the needle with respect to the availability of the more advanced technologies. The November 2010 report also recommended an interagency quadrennial energy review modeled in some respects after the quadrennial defense review.

02:26:42.960 - 02:28:28.960

That has also now been put in train. The first installment of the quadrennial energy review appeared last year. The second installment was rolled out this morning focused on the electricity system from harvesting of raw energy sources to the end uses of electricity across our society and what needs to be done to improve the resilience, the reliability, the efficiency, the economy of the electricity system going forward. Looking out to 2040, the letter report on near term actions on clean energy and climate change that PCAST produced for the president in early 2013 had a profound effect on the thinking that went into the president's climate action plan that was released in late June of that year and continues to be implemented in virtually all of its elements going forward. Of course, as they say, success has many parents while failure is an orphan. I would not want to suggest that PCAST's input on this topic was the only input. There was a very large effort related to the climate action plan that included the Department of Energy, the Environmental Protection Agency, the Department of Interior, and NOAA and NASA and many others. Again, I would argue if you go back and look at the PCAST reports on those topics and then look at what has emerged in this administration's policies, certainly PCAST played a very important catalytic and intellectual role in that development.

02:28:28.960 - 02:29:27.960

As I mentioned also, even though we did at the beginning of this session because it was one of our shorter reports, the influence of the report that Rosina and Maxine and Michael led on increased cooperation between the private sector and the government on preparedness and resilience against climate change. I think the record on that topic is extraordinary. Some other studies that we did that impinge on parts of it, I would mention the study of ecosystem services and how influence on ecosystem services and the valuation of those services needs to be part of government decision making. Making ecosystem services are of course at serious risk from climate change and that is also in a larger sense a part of PCAST contributions to the climate change agenda. In the interest of time, I will stop there.

02:29:27.960 - 02:29:34.960

Well done. You are closing closing remarks. All right. Well done.

02:29:34.960 - 02:30:01.960

I think it's clear just from the conversation we've had that a lot went on. We didn't have time to rehearse and we didn't intend to rehearse every single thing that had come out. As John mentioned, there were 39 reports, 440 recommendations. We didn't want to recite them all, but I think just telling some of the stories of what went on over the last eight years underscore one very important message.

02:30:01.960 - 02:30:55.960

America has been the single place in the world and in history that has best understood this magical interaction between public goods and private goods. And this has informed so much of what PCAST has done that we need an incredibly vibrant private sector in this country. Creating innovative businesses to develop the science and technology and innovation that has given rise to so much of our GDP growth, our wealth, our prosperity over the course of the last century. We've done that better than any other country or any other time in history, not just though, because we have a great private sector, because almost all of it rests on public goods.

02:30:55.960 - 02:31:29.960

That is to say, things that a private enterprise can't themselves invest in and reap the full benefit of and therefore would necessarily under invest in. These include education. Companies aren't going to invest in basic education of their workers because they can't reap that return. They include fundamental research where you can't say where it's going to go and in any case the investor couldn't possibly reap all of the benefits that could come out of it.

02:31:29.960 - 02:32:13.960

They include creating the kinds of common grounds such as this advanced manufacturing efforts or efforts in cities or spectrum, all of which are going to generate huge amounts of economic activity. And the government will make its investments back in the form of the prosperity and the health of its citizens and the taxes they pay that get reinvested in creating that ecosystem. It's a poorly understood thing. We don't do a good enough job of explaining it that the partnership between public and private and the great investment in this country, several percent of our GDP in research and development going on in both public and private,

02:32:13.960 - 02:32:35.960

and the care and tending of that ecosystem is what undergirds our prosperity. I'm very proud to have seen so many different reports that have had at their core that fundamental understanding. It's an important one to pass on. Each administration thinks about how it's going to interact with science and technology.

02:32:35.960 - 02:33:16.960

I think this has been a remarkably broad and bipartisan effort to see this power of science and technology and how it helps and how one has to steward it to get things done. And to then unlock just the creativity of the market of the small startup that's going to make the hearing aids or things like that. So it's been a great honor and a great privilege to serve the president, to serve the nation, to work with all of you on PCAST and on these reports. And I think the reports will have legs far beyond what occurs in these eight years, the next eight years or the next decades ahead.

02:33:16.960 - 02:33:51.960

I think these have ideas that will have long-term impact, and I want to thank everybody for it. And I will turn, as usual, for the last word to my co-chair, John Holtron. Well, practically everything has been said, although not everything has been said by everybody, so I will make just a few further remarks. First, I think that the degree of success and influence that this PCAST has had, which has been extraordinary, was enabled, above all, by the characteristics of our boss, by the characteristics of President Obama,

02:33:51.960 - 02:34:36.960

who was an exceptionally astute and willing enabler and listener and embracer of findings and recommendations about the ways that science technology and innovation could improve our economy, could improve public health, could help protect the environment, could produce a cleaner energy future, could contribute to national and homeland security, his awareness. And he brought that awareness to office. I sometimes inappropriately get credit for having somehow turned him into a geek, a techno nerd. He was one when he came. He had a deep understanding when he arrived at the presidency of how and why science, technology, and innovation matter.

02:34:36.960 - 02:35:44.960

And that made our job as PCAST both an enormous pleasure and an effective influence on public policy because our boss was so prepared to take on board insights and ideas from the wider science, technology, and innovation community. I have said publicly on previous occasions, I'll say again, I think President Obama is the most science savvy president since Thomas Jefferson, with the qualifier that there's a lot more science to be savvy about today than there was when Thomas Jefferson was the president. And that has been, I think, to the great advantage of our nation that we had a leader who had figured this out and who empowered all of us in PCAST, all of us in OSTP, all of the scientists, technologists, engineers, mathematicians, innovators across the federal government. And indeed, as I mentioned before, was prepared to draw on the insights of the full science, technology, and innovation community.

02:35:44.960 - 02:36:49.960

We have had a great run. I have had a great run in my position as his science advisor and co-chair of PCAST. I have had in this group the most amazing set of partners that any presidential science advisor could want, led by the extraordinary Eric Lander, who devoted a fraction of his energies to this operation that is, to me, unimaginable, given the demands of his day job. But somehow he did it. And I think all of us in PCAST recognize that Eric's fingerprints are in one way or another on virtually every one of the 39 studies that we did, and I am enormously grateful, as I know the president is. But again, one must mention the PCAST secretariat, represented here now by Ashley Preditz and Jennifer Michael, and their predecessors who deserve great credit as well, all of OSTP who worked seamlessly with PCAST in its efforts.

02:36:49.960 - 02:37:41.960

And I want to mention as well, Crin Liu and Laura Sosa, who have been doing the logistics work and meeting plans. They have spent a lot of early mornings, a lot of late nights, a lot of last minute requests, always with the smiles that you're seeing now. We are very grateful to them as well. And finally, of course, to the extremely attentive and resilient representatives of the wider community who have attended our meetings and watched them online. We are grateful for your interest as well. And with that, we adjourn the last meeting of President Obama's PCAST. Thank you all.

02:37:49.960 - 02:37:51.960

Thank you.