physically engaged and inspiring young minds as they share their opinions on the issues that are important to them and affect our world. >>■i c-span's washington journa, a live forum involving you to discuss the latest issues in government, policy. from washington, d.c., and across the country. wednesday morning, former transportation department inspector general discusses airline safety, federal oversight of the aviation industry, and multiple safety issues and incidents at boeing. and then an assistant science at the johns hopkins bloomberg school of public health talks about efforts to limit lead and harmfulls in drinking water. c-span's washington journal. join the conversation live at 7:00 eastern wednesday>> c-spand

view of government, funded by these television companies and more, including comcast. >> you think this is just a community center? it is way more than that. >> comcast is partnering with a thousand community centers to create wi-fi enabled -- so students can get what they need to be ready. comcast pports c-span as a public service along with these other television providers, giving you a front row seat to democracy. >> biden administration officials, lawmakers, and advocates discussed health policy issues at politicals health care summit in washington dc. during this session, attendees heard from the chief scientific and medical officer of verily,

primarily focusin■6g on the n of artificial intelligence. this is about 15 minutes. >>t&■b please welcome the host f the political podcast -- the politico tech podcast. i am stephen overly. i'm thrilled to be joined today for a special live taping of the 0;b< dr. andrew tryst are -- andrew trister. he is an oncogwho spent the last decade at the intersection of health care and technology and has had roles at

the gates foundation, and today medical officer at a life science company spun out of google. andrew, thank you for joining us. andrew: thank you. great to be here.t how ai is transforming health care and the need to regulate it, because it seems that there is a lot of excitement and talk aboum ai, and some of those benefits are here, but also some feel, frankly, a bit like hype. so let's start. the company is a precision health company, which i understand it, is care tailored to a patient in their individual oedipal history, environment, and needs.p what is the most significant way you are seeing the way ai is transferring the way patients are looked at right now? andrew: even though there is hype, and that is true since chatgpt was released last fall, the ct working on ai and medical care for

decades, and that originally was in the forof dr. make a better decision, expert systems in the 1970's. it did not work very well because the totality of the data and the ability to take what we

understand as physicians interacting with doctors was not ther-e. later we move towards augmenting our ability to diagnose, so looking at cells under slides for pathology, looking at radiology and images from ct or chest x-rays. those technologies have been regulated, and there are hundreds in the hospital today impacting care, but i think where the hype comes in comes into can we get beyond those nd the experience of the patient and push into what does that patient see, so how can they access information generated by him or about them in a different way. the health care behind the scenes before the patient is necessarily involved in what they actually interact with. when i talk to folks about ai d health care one of the regular topics they bring up his drug development and the potential for ai to speed up breakthroughs in medical discoveries. i guess one question i have is to what extent is that real today versus something that we are looking to or in the future that there is this potential for an innovn medical breakthroughs, clinical trials, all of that happens at a faster pace? andrew: this is an area of intense interest and certainly has been accelerating over the last couple of years. so there has been lot ofcompanie at alphabet, particularly from deep mind, one of the ai groups that got pulled into alphabet as an acquisition and has spun out

something called isomorphic, whis the protein structures, something that has eluded scientists for a long time.■i nt and the question will be from these proteins, can we figure out where to drop them? we can build the drug more easily. they are just one example of a few other companies that do the same thing. that does not mean it is a medicine, though. steven: right.sv■! dr. trister: that is the part that has been haer. this for a long time haven't yet gotten to a regulatory filing and of medicine available on the market, so we are still waiting to see can it still speed things up, but it is promising. steven: is that the big hurdle still to be achieved, taking this and making real drugs that help people?■k dr. trister: absolutely. the science is there. that does not mean that is tolerable or safe, so going through what is now the pathway that fda have laid out and other

regulato that drugs or the potential medicine is safe and effective,t we still need to take. steven: speaking of the fda, i want to talk about the regular -- the ai. in particular, ai is only as good as the data driving it. you have to acquire a lot of data, and in the case of health care, a lot of personal data, and companies like verily are being entrusted with the information, some protected by hepa and other privacy regulation, but some not, so i'm wondering what regulation might be missing that is needed as ai becomes the norm? great question. certainly after the biden white house put forwardhe exetider ont was showing ai touches every aspect of life, and there is the

potential for us as a society to recognize where is the most individual and how is it we might be able to protect this, but not as a hurdle getting us to that utopia where perhallctos and have better diagnosis, faster care, cheaper care. that is the promise that we bee ai can get us to. steven: the vision we are working towards. dr. trister: right, but to get there, we have to understand where people feel comfortable today, and understanding what data could they contribute, what data can they generate, and what is done already within th health -- within the health system and what they can do to have control over the data. that is not enshrined in data like hipaa, even though it did

not anticipate the data revolution, but i think there are opportunities to update and demonstrate what is good for the society and how we still protect individual rights around control particul what data and what it is used for. steven: how much of the responsibility is to come up with those rules like a company that discloses how the data used and that sort of thing? dr. trister: everybody would like to see regulations are clear. we want to take steps -- sensitive data about a person so weis this is how the data can be used and how we will not use it. we will never sell it, as an example. those are reasonable guidelines all companies that work and health care should follow but that has to be enshrined in regulation. steven: continuing with regulation, washington is in the eastandards need to be set around ai. there is this coalition of health ai, industry and academi-

assurance labs, setting some of these standards in the health carepa parent company, is part of the coalition, so one question i'm left with is, what is the appropriate role for private sector companies that have a financial interest at that will govern our health care data? will andrew: i think the coalition for health ai is a great example where private industry has come together to recognize the nes for patien■1i■ advocacy for providers and government officials. there are multiple stakeholders when it comes to any health care regulation, and the goal, we will see how it comes about, but with the coalition for health ai, the goal will be to allow each stakeholder to have a voice and to impact how these regulations can come about, but these assurance labs are a good first step. we don't have to rely upon fda or other government agencies to

keep up with all of thanre happ. ai is moving at such a clip at the moment, it would be really hard to imagine anyone, even those people working on it, being able to keep ahead, so instead the assurance labs can be one of those areas, a crucible to test what es can come about. steven: when you talk to tech folks about innovation, there id break the things. 're about the data, the study, what we are talking about at the conference, two different paces ofovhow do you reconcile that, and do fall into one camp or the other? andrew: there is a huge tension in my brain because of that. i fall more onhe showing data, but i also recognize adoption within the health system today is glacial.s

been almost generational. you have to wait for the new docs trained with the technology to start using it, and that is not appropriate. there are people who today we can affect their lives and they can live better, longer, healthier lives with technologies that exist today. just the gatekeeping elements of how the health system is set up reduces the applicability so that there is an equitable question about how people access these tools, and sometimes that is the health system itself, the actors within it, the physicians and payers, so that could be regulated as well. steven: that is an interestingvr powerful it is, whatever brntegrated into an existing system that we know in many ways is broken or could be improved. i want to ask about a couple of those.

one is the cost question, because when you talk to people about health care, cost concerns is one of the first things they talk about. is ai going to make health care cheaper? more expensive?tely be something that really benefits the wealthy? andrew: i think that is a reasonable concern and absolutely something we have to recognize, again, as a society and as groups that are going to make these decisause the likelihood is that many things will follow where the money is. that is the rule of a free market economy. we heard earlier today about the role drugs play in oncology. oncology alone, care and research together, are 1% of gdp in the u.s., not 1% of health care spend, 1% of the economy is oncology care and research, so that drives a lot of innovation towards oncology, and as an

it on. this is great. steven: right. dr. trister: but that does not necessarily apply in the same equitable way as other conditions like alzheimer's or gop ones. and we talked about how is it as both regulators and payers, wham ensuring there is access and that it is equitable for full to be able to get these new innovations. who in the health care continuum should bear those costs? there are going to be some health-care systems tha able to invest in ai and new technologies, and some, particularly smaller or rural how do you bridge those divides? dr. trister: right now, u.s., which we saw over the last four years, actually on the anniversary of the global pandemic being declared march 11, that in that

time, what were little fissures, and we recognize there were socialss and carrier delivery, those became massive chasms. art of that was the misinformation and disinformation we heard about earlier, but imathink the infra, the underlying technology that allows things to flow from a patient in their home, in care,o be a public good. once that exists, build all th. but right now, there are tools being built within systems that onlyenalready have access to the health system, because they are built on the data of people who already have access to the health system. it is data already there. hink about the people who could not even get to a doctor's appointment or could not pay for the drug to be included in the data set, we

as to that population. steven: the other piece of this, i wonder to what extent our health care system is ai and pl this data. we have talked about this a little bit, but i think of doctors offices i go to where i fill out a and i show up and they hand me the same form to fill out on paper. there is a dichotomy about how we talklth care and how it is actually delivered in practice. to what extent are providers equipped to even handle this sort of new■av everything that ai is going to throw at them? dr. trister: the way the technologies have been integrated into the clinic have not taken into account what it the electronic health record systems that we have become a now in thence the american recovery act kind of insured that digitization of the health care system would occur

did not anticipate what the experience would be like if your physician is spending more time starg at a screen and typing rather than looking at you and asking questions. that has eroded trust in the health system. that has affected manymy colleagues at a level of burnout, because there is much more time spent coding and billing rather than actually caring for people, the whole reason people go into medicine at the end of the day. i think ai plays a role in to automating some things that are truly mechanical, like billing and coding is an example. there are many examples now of many other startups and companies that are doing the sameat those get integrated appropriately with the right security and privacy around the data, th will not have to spend so much time on it. the issue is that substrate is not the right one. put it on topth that dhr has bea

barrier to care instead of an enabler. dr. trister: our time is -- steven: our time is about out, but a lot of questions about the solutions that need to be applied to the health care system before ai can have the benefi■óaldr. andrew trister, tr you get your podcasts. i will also be in the lobby doing interviews, so feel free to stop by and share your opinions on ai. but thank you for being here. >> discussion on the israel-hamas war after nearly six months of fighting, and a look at the prospect of a cease-fire. watch live at noon on c-span, c-span now, our free mobile video app, or online at c-span.org.