{"url": "https://tracking.swap.fm/track/IKyd32SdEoqWH6LmBFTJ/podtrac.com/pts/redirect.mp3/washpost.pdn.tritondigital.com/posttv/washpost-production/5a21a2d9e4b0f49da26f53cf/20180330/5abe6296e4b00013b918d82e/5abe6296e4b05fcebe3f4106_1351620000001-300030_t_1522426521420_44100_160_2.mp3?awCollectionId=5a21a2d9e4b0f49da26f53cf&awEpisodeId=5abe6296e4b00013b918d82e", "page_url": "https://www.washingtonpost.com/podcasts/post-live/transformers-artificial-intelligence-people-robots-and-society/?tid=aud_rsslink&utm_source=podcasts&utm_medium=referral&utm_campaign=post-live", "title": "Transformers Artificial Intelligence: People, robots and society", "show_or_channel": "Washington Post Live", "date": "2018-03-20", "duration_s": 1972, "type": "podcast", "confidence": "high", "notes": "Panel with Milind Tambe and Meredith Whittaker; Clark at OpenAI", "src": "p3", "id": "2018-03-20_washington-post-live-transformers-artificial-intelligence-pe", "audio_duration_s": 2031.97825, "worker": "jc-w08", "transcribe_s": 86.6, "model": "large-v3", "segments": [{"start": 0.0, "end": 22.1, "text": "Get The Daily Show's ongoing coverage in Decision 2026 daily. We're committed to being accurate. It's the most trusted fake news in America. Comedy Central's The Daily Show. New weeknights at 11 on Comedy Central. You're listening to a podcast from Washington Post Live. Bringing the Post's newsroom to life on stage."}, {"start": 26.16, "end": 51.07, "text": "On Tuesday, March 20th, the Washington Post brought together pioneering researchers, business leaders, and elected officials for Transformers Artificial Intelligence, this live news event focused on advances in technology that are poised to reshape the way we live and work. In this segment, Jack Clark, Strategy and Communications Director at OpenAI, Malind Tambay,"}, {"start": 51.07, "end": 73.65, "text": "founding co-director of the Center for Artificial Intelligence in Society at the University of Southern California, and Meredith Whitaker, the co-founder and executive director of AI Now Institute, sat down with The Washington Post's Jeremy Gilbert to examine the ethical and societal implications of artificial intelligence. Let's listen."}, {"start": 81.33, "end": 109.1, "text": "All right. Good morning. I'm Jeremy Gilbert. I'm the director of strategic initiatives here at the Washington Post, and I'm thrilled to moderate our last panel of the morning on the ethical and social implications of artificial intelligence. I'm joined by a real esteemed panel of experts. Jack Clark is the director of communications and strategy at OpenAI, which is a nonprofit AI research company. Meredith Whitaker is the co-founder and executive director of the AI Now Institute at NYU University."}, {"start": 109.17, "end": 133.12, "text": "And Milan Tambe is the founding co-director of the Center for Artificial Intelligence and Society and the director of TeamCore Research Group on Artificial Intelligence and Multi-Agent Systems at the University of Southern California. Thank you all for being here. If you have any questions for our speakers, please tweet them to us using the hashtag Transformers, and I'm going to get the discussion started right now. So this is an interesting group."}, {"start": 133.12, "end": 156.86, "text": "With some very different perspectives. But one of the things that seems to unite you is the importance of the impact of artificial intelligence. The kind of question of how we weigh, for example, what large institutions might have as business interests using artificial intelligence versus societal good. How do we decide?"}, {"start": 156.86, "end": 185.78, "text": "Who gets to be the beneficiary of the gains of artificial intelligence when we're weighing those two things? So you're starting with the easiest question. Of course. Well, I guess I'll start and we'll go down. But OpenAI was founded a little over two years ago. And the goal of OpenAI is to ensure that the benefits of advanced AI accrue sort of widely to all of humanity rather than just a few. And I think that everyone on this stage shares that idea."}, {"start": 185.78, "end": 214.3, "text": "It's interesting to me that such an organization was formed at this time because it suggests that there's huge anxiety that it's going to happen in the other way. And we're not going to see those benefits widely distributed. And I think that there's a responsibility for the AI community centers to try and play a larger role in the kind of governance and the defining of norms around this technology so that it can go well. And there's huge anxiety that the default setting"}, {"start": 214.3, "end": 233.76, "text": "is currently one where the private industry gets to define the rules of something which affects everyone, which I think we should be somewhat nervous about. Yeah, I would hard agree on that. The AI Now Institute at NYU, I co-founded that with Kate Crawford in part because we want to look at..."}, {"start": 233.76, "end": 254.22, "text": "What is AI doing now? What are the impacts that are happening right now as early AI systems are being rolled into the infrastructures of our daily lives? So in response to this question, I want to bring it back down to the practical. I would agree with what Jack said. But at this point, we don't even have an accounting."}, {"start": 254.22, "end": 284.08, "text": "for where these systems are integrated into the back ends of core decisions. And this is why the AI Now Institute called for an elimination of black box systems. So this is sort of unaccountable, obscure systems not subject to oversight as used in core government agencies. How many of these black box systems are there? What's the scale like? We don't know. Like, full stop. It's a fundamental challenge to accountability."}, {"start": 284.08, "end": 311.9, "text": "For where these are being used, for what they are, is this an overburdened spreadsheet or is this a neural net being applied to determining somebody's Medicaid disbursements, right? So where we do see problems, and we see them frequently from policing heat maps to the case I'm referring to in Idaho where Medicaid disbursements dropped by 30% and people didn't know why because they didn't know how the system worked, right? A number of these cases crop up. These are kind of the tips of the iceberg."}, {"start": 311.9, "end": 337.73, "text": "This is where investigative journalists or where researchers are able to get access, are able to get information, where a whistleblower, say, comes forward and talks about these systems breaking down in a specific way. But we're in a situation where these are making determinations about people's access and opportunity oftentimes, or most times, I would say, without the individuals who are affected even knowing that the system had a role. So auditing and accountability are..."}, {"start": 337.73, "end": 366.24, "text": "These are core issues that we need to first build the social framework to accommodate, I would say, before we continue rolling these systems into core decision making. And this is AI Now just published an algorithmic impact assessment framework for the New York City Algorithmic Accountability Bill Task Force, kind of beginning to put some structure around these ideas and suggest ways of starting to audit and assess and account for these systems."}, {"start": 366.24, "end": 395.31, "text": "The University of Southern California Center for Artificial Intelligence in Society is something I co-founded with Professor Eric Rice. This is a collaboration between AI researchers and social work. And so we're very proud of this sort of collaboration that's interdisciplinary, but really with social workers who are out there in the field. And so the grand vision is AI to..."}, {"start": 395.31, "end": 423.76, "text": "Address the grand challenges of the American Academy of Social Work and Social Welfare, ending homelessness, achieving equal opportunity and justice, and so forth. There are 12 grand challenges or grand challenges of the National Academy of Engineering. So within those, I guess our focus has been on concrete problems where we can really assist and make a difference in working with low-resource populations, such as with homeless youth, or in conservation, trying to protect endangered wildlife."}, {"start": 423.76, "end": 446.48, "text": "It sounds like this is a very impact-focused kind of artificial intelligence."}, {"start": 446.48, "end": 476.08, "text": "Do you feel like the field at large is focused on impact, even whether it's business goals or social good, or do we run the risk of AI as really working towards novelty, trying to solve an interesting but not useful problem? So this is a question that's very near and dear to my heart. Speaking to the research community, I feel that in AI, we need to focus more on impact."}, {"start": 476.08, "end": 504.22, "text": "You know, when we publish and so forth, novelty is given a higher weight and impact is not. And so as a result, but it's up to us as researchers, as senior researchers in AI, to redefine what are important ways to measure progress for younger researchers. Because if the reward system is such that what counts is novelty and not impact, then essentially that's where people will go. But impact, and if you want to see societally beneficial impact."}, {"start": 504.22, "end": 534.05, "text": "Then that's something that we as researchers need to define as an evaluation criteria so that people speak to that and do research in that area. I'd just like to jump in on this quickly with a tangible example. So along with OpenAI, I also do a project called the AI Index, which is about tracking progress in AI. And what you discover there is there's this premium placed on novelty that means that we think AI is progressing in some way faster than it is. And I'll give you a specific..."}, {"start": 534.05, "end": 560.61, "text": "Kind of specific example, last November, Alibaba and Microsoft claimed they'd reached human performance on question answering on a Stanford data set called Squad. You don't need to know too much about it. I remember the fact that they issued press releases saying they'd reached human level and machines could now understand paragraphs of text as well as a human. Everyone got very excited and thought, well, I guess that's done now. We can move on to other stuff."}, {"start": 560.61, "end": 587.68, "text": "And the Allen AI Institute, which is from Microsoft co-founder Paul Allen, just released a new data set called Arc, which also tests common sense reasoning over natural language. And every single existing technique, including the ones that do well on Squad, totally fail to do anything interesting on Arc. And their performance is, maybe they get it right 25% of the time. So that's subhuman. Human level is meant to be 90%. And so I think that by..."}, {"start": 587.68, "end": 616.27, "text": "Prioritizing like the new, new thing, you risk making us think that A, AI is progressing faster than it is, and B, it sort of subverts the effect of doing useful stuff that actually works in the real world. And I think it's important, especially for you, you know, in your position here at the Washington Post, to always be pushing companies to say, but what does it actually do that's useful? And I guess we're not saying that novelty has no value. Obviously, you're saying compliment or"}, {"start": 616.27, "end": 643.52, "text": "You know, add additional weight on social impact so there's more balance and not just sort of pure novelty. And there is research going on, but I feel not enough. And we could do more to encourage that. Please. Just a quick addition there. I would agree with both Milan and Jack around sort of novelty driving a kind of warped understanding of AI progress. I do maybe want to..."}, {"start": 643.52, "end": 669.2, "text": "Take it back to this term impact and pick that apart a little bit because I actually see the recent AI boom being a recognition that you could impact bottom lines by commercializing this technology, right? And there's a lot of people who've been working since around 2012, but sort of an increasing exponential arc to figure out how do we market AI products. So when I talk about black boxes deployed in core government agencies,"}, {"start": 669.2, "end": 698.48, "text": "Chain of title there is often to vendors who are selling extreme promises based on the capabilities of systems that probably haven't been tested in the populations they're going to impact. So there's an issue here, I think, not with AI as having no impact, but with the impact as being measured by bottom lines in sleek tech conference rooms. And actually, there's kind of an air gap between that and the environments that are actually being significantly shaped by the determinations of these systems at scale."}, {"start": 698.48, "end": 726.1, "text": "So how do we close that gap? How do we get from saying there are specific researchers or engineers who are designing these algorithms that they are then either placing in a black box or maybe in some cases exposing to having a voice from the people who are being impacted by those algorithms? What's the way that we close that? So at least for us, one major way of trying to do this is through immersion."}, {"start": 726.1, "end": 742.18, "text": "So our students, myself or others, we actually want to go to the location where the impact is to be measured and work with the people there. It's problem-centered. It's coming from the domain. And so when it comes to conservation, we've actually patrolled."}, {"start": 742.18, "end": 771.5, "text": "In forests in Malaysia and so forth, really understand what is really the problem or go to the forest in Uganda with working with the homeless shelters in Los Angeles. So be there. We are working with social workers. It's fundamentally interdisciplinary. It just can't be AI researchers figuring this all out by ourselves. And I always try to think about showing humility, so working with public safety or security, working with the Coast Guard, for example."}, {"start": 771.5, "end": 796.59, "text": "We can't be sitting in LA at USC and just saying, we know better how to drive your boat, so we're going to tell you. We really need to go beyond the boats in New York and understand how they work so that we can develop algorithms that are more appropriate to the location and are done in an interdisciplinary fashion, are responsive to the needs of those who are using it, and so on."}, {"start": 797.65, "end": 820.96, "text": "Is there also a question of, you touched on Meredith transparency, but also regulation. So who gets to say whether these, the algorithms being used, the automated approach being selected is fair and fair to whom?"}, {"start": 820.96, "end": 848.11, "text": "Yes, this is the multi-trillion dollar question. I think we do need shared standards across industry. I think regulation could absolutely be helpful in some of these cases. I think we need to have a clear-eyed understanding that we need to validate before we try in high-stakes domains. Now, we're in a situation where the acceleration of commercialized tech has had resources behind it for a couple of decades."}, {"start": 848.11, "end": 876.26, "text": "We are not in the same situation where the research field that is looking at the nuances of measuring lived impact across dynamic contextual domains has had the same amount of acceleration. So there's a real need to center fundamental research on these questions. There's a real need to, I think, transparency, I would say, is part of the equation. We need to know where these are. We need to know the justifications for their use. We need to know by which standards they were audited. And we need to have"}, {"start": 876.26, "end": 903.25, "text": "clear democratic processes to either accept or reject that use over time. And that's part of what we propose in the algorithmic impact assessment framework, which is being led by our law and policy research team. And this is an area where I think government has a clear role to play. Like, you know, earlier today we heard from the senators and we regularly hear from policymakers about how they want to do something about AI. I think measuring and evaluating..."}, {"start": 903.25, "end": 928.46, "text": "And assessing how AI is deployed is what government should be in the business of doing. I would like the government to tell me that it has tested all of the autonomous cars on the road and could give me reports on their performance. And that's not based on the tragic accident that happened yesterday. This is a long-term problem. Like if I want to understand self-driving car progression today, I need to go and look at the reports filed with the California DMV."}, {"start": 928.46, "end": 952.13, "text": "Which tell me the number of disengagements that each autonomous vehicle makes per vendor. And then I need to build my own spreadsheet and then I get a view of performance in that state. All of the companies have since moved their testing out of California because they don't like me being able to find this out. So actually now it's spread around the country and no one knows. And this is terrible. This is like a technology that is going to dramatically affect our economy, affect people's lives."}, {"start": 952.13, "end": 976.61, "text": "And we aren't assessing the rate of progression in it because the private sector is saying, well, that's too commercially sensitive. You can't really do that, which is absurd. So I think that having government take more of a role here would be beneficial. And it would also force information into the public domain, which isn't. Like, as Meredith said earlier, if I want to find out about where government is using"}, {"start": 976.61, "end": 1003.33, "text": "AI today, I need to do freedom of information at requests, or I need to work with the ACLU, or I need to call up people I know in government and just go and have a beer and ask them about what machine learning system broke today. None of that feels particularly healthy. So yeah, I think this is a pretty serious issue. Well, and the question of, for example, different states having different standards is particularly interesting, but AI has been applied."}, {"start": 1003.33, "end": 1023.78, "text": "Internationally, on what level are we talking about? Are we talking about a municipal level, a state level, a federal level, or an international level? Do we need international standards? All of them. I'll offer a quick comment and then turn it over. I think we're oftentimes talking about massive global systems at a huge scale."}, {"start": 1023.78, "end": 1051.22, "text": "So as we've seen with GDPR, as we've seen with other regulations, the high watermark for regulatory intervention matters because you're not going to bespoke that system a number of times. So I think all of them make sense because contexts vary, norms vary, et cetera. But we should attend to the power of smart regulatory intervention. So I think, I mean, I agree with all that was said earlier with respect to"}, {"start": 1051.22, "end": 1071.44, "text": "You know, government help in regulation and so forth. But in addition to that, as we discussed earlier, it's also within the research community. This is now sort of evolving into more of an interdisciplinary science, and therefore the need for us as researchers in AI to reach out to other disciplines."}, {"start": 1071.44, "end": 1095.63, "text": "People in other disciplines. And really to force ourselves to encourage ourselves to do these kinds of measurements. And I'm in full support of what Meredith was saying earlier with respect to really doing a better job of measuring things in the field and really measuring the impact and assessment. But these sorts of things today are difficult because if I'm a researcher,"}, {"start": 1095.63, "end": 1123.31, "text": "Publishing these sorts of things, getting encouragement from the impact and so forth is a little bit harder. So there is a role for government, but there's also a role for AI researchers to do something about this. So specifically for the AI Index, which does this measurement initiative, we were trying to find a new person recently who can join the kind of core team, which includes myself, that works on it. And we were talking to a really smart..."}, {"start": 1123.31, "end": 1152.59, "text": "Young pre-tenure track professor, or about to be professor. We're saying, do you want to join the index? It's a chance to do something with some impact, and it's a chance to set norms. And they said, no, because I'm pre-tenure. So this would not be evaluated positively. I need to be doing technical contributions. And that is kind of frightening to me, that the incentives are set up in such a way that academics who want to work on impact find it challenging to."}, {"start": 1152.66, "end": 1182.66, "text": "We have a little trope we throw around AI Now to explain why we are so constitutively interdisciplinary across six faculty at NYU. And it goes a little something like you wouldn't expect a doctor to tune a deep neural net. You shouldn't expect a computer scientist to make complex decisions in fields like medicine, et cetera. So we're really looking at this sort of expert drift where in the computer science field, which sort of dominates"}, {"start": 1182.66, "end": 1212.43, "text": "The big tech industry that is driving a lot of this innovation has sort of taken it upon itself. And I think many people are deeply uncomfortable with this within and outside of these companies to make really significant decisions that affect domains that are outside their realm of expertise. So I think we need to look at a fundamental restructuring of what we call product development, of what we call research, of how we define AI research to include sociologists, ethnographers, legal scholars, et cetera, et cetera."}, {"start": 1212.43, "end": 1232.98, "text": "How are those decisions getting made? Are the computer scientists identifying problems they want to solve outside of their own disciplines? Are people coming to them? How do you make that a more inclusive process? So I think what should happen, and to some extent we strive to make it happen in our center."}, {"start": 1232.98, "end": 1261.44, "text": "As I was saying earlier, user-centered problem solving. So you start from the problem. And for some people, it's kind of confusing. Why do you start from a problem? Why not start from a solution, which is not necessarily the best approach? But anyhow, so I can give you kind of this concrete example where we were trying to come up with better patrolling methods."}, {"start": 1261.44, "end": 1285.39, "text": "For protecting endangered wildlife in Malaysia, for example. And we could sit in LA and come up with, oh, this is the best patrolling route. And the people that we were talking to on Skype were saying, no, no, this doesn't work. This absolutely doesn't work. You know, the shortest distance between two points is not a straight line. And to us, it's like, how can that be? So we flew down to Malaysia. We actually patrolled in the forest."}, {"start": 1285.39, "end": 1308.3, "text": "And suddenly we realized, yeah, you have to walk along ridgelines. You have to walk along riverbeds. You can't just walk in a straight line in a forest to patrol. It doesn't make sense. These are types of things that can only be gained by being on the spot in real life rather than just sitting in the lab and saying this is a... Now, I want to be sure..."}, {"start": 1308.3, "end": 1337.23, "text": "There's a lot of wonderful theoretical work researchers are doing. This must continue. I mean, pure basic AI research has to continue. But there's space for this additional kind of work. And this requires us to get out of the lab and get out in the field. And rather than focusing on what is readily available as a data set, like ad auctions and so forth, to focus on new kinds of problems where data sets are not necessarily easily available."}, {"start": 1337.23, "end": 1363.36, "text": "Where it is harder to do that kind of work. And some of it is about changing norms so that researchers think that they should intentionally be multidisciplinary. At least for some projects. We did a project recently on malicious actors in AI. So how you expect really unpleasant people to take open source AI technology and do unpleasant things with it. And so to do that, we ended up hosting a workshop."}, {"start": 1363.36, "end": 1387.12, "text": "About a year ago in the UK, and we had people from the police come along, people from intelligence agencies, people from the AI research community, sort of black hat hackers who had in a past life been one of these nasty people using open source technology to do unpleasant things. But it was very helpful because then you're there with people with a huge spread of skills."}, {"start": 1387.12, "end": 1413.63, "text": "Outside of this narrow technical domain, who can tell you, no, this is the real problem, or this should be your real threat model. And I think the more we can do with that, the better. And AI now has been sort of leading some of these initiatives already. I would agree. I think also we need to, I've been in the tech industry for over 11 years now, so I'm what they call a veteran."}, {"start": 1413.63, "end": 1427.33, "text": "And I think we really need to examine the culture of tech because I will testify from my own experience, it feels like it's gotten less diverse and more homogeneous as its sort of power has ascended."}, {"start": 1427.33, "end": 1446.3, "text": "And I think, as Jack was saying, you can have the best intentions. But if you're in the room with people who fit a very homogeneous demographic, who have shared the same experiences, whose problems involve the laundry delivery maybe didn't come on time that day, you shouldn't be expected to have this sort of infinite."}, {"start": 1446.3, "end": 1475.95, "text": "Thank you so much for being here."}, {"start": 1475.95, "end": 1502.54, "text": "I mean, this is essentially part of the challenges. If the people you put in the room are codifying their own biases into the algorithms that they create, you're going to A, get back the results that they probably expect, but B, probably not recognize all the problems."}, {"start": 1502.54, "end": 1532.08, "text": "Inclusion is certainly one technique. Is there an issue of awareness as well? Yes. Yes. And there's many other issues, right? We're talking about one issue that is sort of problematic. There's issue with not examining the data sets you use, right? Are you using data collected by the Baltimore Police Department's Gun Tracking Task Force, which is now under investigation for criminal conspiracy, planting drugs on suspect? This is one of the largest police..."}, {"start": 1532.08, "end": 1561.47, "text": "misconduct cases in the country, are we then deploying predictive policing that would use that data as ground truth to determine who looks like a criminal or not, right? There's a lot more investigation we need to do on what are the fundamental claims being made in the data, what is the history, who builds the classifiers and the algorithm, what is the viewpoint that is imparted there, then what is the context in which this would use and what are the sort of power asymmetries and other issues that might attend sort of, you know, a cop getting a score."}, {"start": 1561.47, "end": 1583.47, "text": "So I guess in our work, the focus is more on augmenting humans who are making decisions in very complicated situations. So if you're trying to spread information about HIV prevention, which is one of the problems we're working on, who are the right youth to select in a homeless shelter?"}, {"start": 1583.47, "end": 1610.64, "text": "to spread that information. This is a task where you're looking at the social network, trying to figure out who are the right people. These kinds of tasks are already being done by human beings. They're augmenting their capabilities. But we want to achieve the right kind of balance, not over-prescribed. So the right level of autonomy. If you are trying to tell rangers in Uganda that this is where you're going to find snares, well, you can give them a certain area."}, {"start": 1610.64, "end": 1621.41, "text": "This is where you're going to find it. We don't have the ability to tell them, well, you're going to find it under this tree. I mean, they are experts in that domain."}, {"start": 1621.41, "end": 1643.22, "text": "Harness their expertise where possible and leave AI to do what it knows best. And understanding this right level of balance between the teamwork between humans and AI is also an important aspect of research as we go forward. And I think this is from the sort of what we've been talking about, about the narrowness of AI community sometimes. I frequently have conversations with..."}, {"start": 1643.22, "end": 1670.56, "text": "You know, perfectly nice people who are talking about an AI research project and they're like, we must remove the human from this entire process. Well, that's dubious. And I think it's because we aren't developing enough respect for the fact that people are like incredibly smart and incredibly good at doing lots of things. And the temptation within AI is to automate the entire process, whereas you're probably going to have something much smarter in the aggregate if you find a way to."}, {"start": 1670.56, "end": 1700.74, "text": "Have the person use their skills and just remove some of the dull or tedious work. And 90% of the time, that seems to be the better course of action. So the last question I would pose to you, and I think you were very much hinting at it here, is if there is fear in the community, and you wrote about this recently, and by community, I mean outside of the AI community, honestly, it's at this sort of replacement of humans for specific tasks. How do you balance the fact that"}, {"start": 1700.74, "end": 1721.86, "text": "Work and a sense of purpose is really important for lots of people against the idea that there are lots of rote tasks and things that can be done to either augment or to avoid certain actions. Where's the balance? I would start historically and just mention that before the..."}, {"start": 1721.86, "end": 1735.44, "text": "You know, mid-19th century and the Industrial Revolution, we really didn't have the same structure of work that we have now in the West. So the nuclear family with somebody who went out to do wage work at a factory and came back and the household as this sort of"}, {"start": 1735.44, "end": 1764.74, "text": "Separate realm was not the norm. You had family economies. People participated in productive labor. There was a blending between life and work. And these distinctions didn't apply. So that's one example. We have across cultures and histories many, many, many different examples of what meaningful productivity and interaction looks like for human beings. So I might push back a little bit against the truism that without a timesheet to clock in with, we will be adrift with no meaning. And I would also then"}, {"start": 1764.74, "end": 1790.0, "text": "I think we're a ways from total replacement. I think in keeping with the spirit of AI now, we should look at what's already happening. We have AI driven hiring. We have AI driven management. We have precarity economies like Uber and other things that effectively run large machine learning systems to"}, {"start": 1790.0, "end": 1819.73, "text": "So what are the impacts of these small encroachments? What are the power asymmetries that are emphasized there? And how do we draw on what we know today to understand possible trajectories and not look for this bright line between humans with work and humans adrift without meaning? I just want to pile on to what Meredith said quickly, which is in two specific points. One is that we need to broaden our definition of work."}, {"start": 1819.95, "end": 1848.8, "text": "Emotional labor is a significant amount of work that is broadly kind of uncompensated across society. We have like an aging population who need like good social relationships. I would like to go and, maybe I'm strange, but I'd like to go and get the chance to go and talk to people and have that be seen as my work, just along with my work at OpenAI. And, you know, there's lots of jobs that people would like to be paid for, but we just choose not to compensate the society currently. And the second point is,"}, {"start": 1848.8, "end": 1868.29, "text": "And I think Meredith touched on this, but where AI gets deployed into sort of middle class or lower class jobs, there's not much evidence that it makes those jobs more pleasant. It actually seems like it makes those jobs unpleasant. And I think that we should remember that because it's very easy to go to these conferences and think like AI is making"}, {"start": 1868.29, "end": 1896.32, "text": "You know, tremendous strides in productivity, but it's usually doing that at the cost of a sense of human agency. And I think that we can be really complacent in the tech industry about this, but eventually that's going to, like, come home. And, you know, the previous ways this came home were, like, the Luddites, which was a reaction to the Industrial Revolution being mostly terrible during the time it occurred. And also the French Revolution, which was, like, fairly unpleasant for the people that have been complacent about things."}, {"start": 1896.32, "end": 1907.33, "text": "So let's not kind of sit in our laurels and say everything's fine. So I want to agree with a lot of what's said. But I also want to say that there's a lot of"}, {"start": 1907.33, "end": 1933.6, "text": "The immediate benefit that we can accrue from AI by deploying them in low-resource communities for domains like suicide prevention or substance abuse prevention or domains that are important for low-resource communities, for conservation, for public safety. These are things we can do today. And we should not lose sight of the benefits that AI is providing there already."}, {"start": 1933.6, "end": 1967.82, "text": "You know, not get so fearful that we will stop that kind of work. So that is important for us to continue forward. Thank you so much. Unfortunately, that's all the time we have for today. I'd really like to thank Jack Clark, Meredith Whitaker, and Milan Tambay for joining us. If you'd like to watch video clips from any of today's discussions or past Washington Post Live programs, please head over to WashingtonPostLive.com. I'm Jeremy Gilbert. Thank you so much for watching today. Thanks for listening."}, {"start": 1967.82, "end": 2007.04, "text": "To hear more interviews from this series and other Washington Post Live programs, visit us at WashingtonPostLive.com. Looking for the best of Vrbo? Discover vacation rentals of the year, top quality homes guests love from across the country. Our curated list takes the guesswork out of planning so you can spend less time searching and more time packing. Book your next day now with Vrbo. Day or night, Vrbo Care is here 24-7."}, {"start": 2007.04, "end": 2029.33, "text": "To help make every part of your stay seamless. If anything comes up or you simply need a little guidance, support is ready whenever you reach out. From the moment you book to the moment you head home, we're here to help things run smoothly. Because a great trip starts with the right support. And hey, a good playlist doesn't hurt either."}]}