Slate Money

Justin Hendrix and guests discuss AI safety incentives, regulatory gaps, and rising public opposition to data centers.

42 min Listen to the original Episode page
AI SafetyTech PolicyData CentersPrivacyRegulationMetaOpenAIAGI

Summary

Host Elizabeth Spiers interviews Justin Hendrix, CEO of Tech Policy Press, about the conflict between AI safety and competitive market pressures. Hendrix argues that US privacy legislation is fragmented and ineffective, while legislative proposals like the 'Great American AI Act' often miss the mark. The conversation highlights privacy risks in Meta's 'Muse' AI agent and the physical salience of data centers driving local political opposition. Guests note that tech policy is becoming a mainstream 'kitchen table' issue, with researchers warning about recursive self-improvement and the ambiguity of AGI claims.

Key takeaways

  1. 1
    Justin Hendrix identifies as CEO and editor of Tech Policy Press, an independent nonprofit publisher focusing on tech and democracy issues, supported by a global network of fellows and volunteer contributors.

    Unidentified speaker (speaker_3) · 1:58

Episode sections

  1. 0:00 – 1:47

    Sponsor Reads

    BetMGM, WestJet RBC MasterCard, and Optimize Wealth Management advertisements.

  2. 1:47 – 7:41

    Introduction and AI Safety Incentives

    Host Elizabeth Spiers introduces guest Justin Hendrix, CEO of Tech Policy Press. They discuss the conflict between AI safety concerns and the competitive incentives driving rapid deployment, noting that CEOs have agency to slow down but often choose otherwise due to market pressures.

  3. 7:41 – 14:45

    Regulatory Approaches and Legislative Landscape

    Hendrix discusses potential regulatory paths, including new laws versus liability-based approaches. He highlights the lack of comprehensive US privacy legislation and mentions specific legislative proposals like the Obernulti Trehan bill and the Great American AI Act, while criticizing 'kill switch' bills as technically unfeasible.

  4. 14:45 – 20:33

    Political Motivations and AI Dominance

    The speakers analyze President Trump's support for AI, linking it to economic dependence on AI investment and a personal obsession with dominance. Hendrix expresses concern over the concentration of power in a few tech companies and the blurring lines between government and tech firms.

  5. 20:33 – 26:45

    Meta's Muse and Privacy Risks

    Hendrix details privacy and security risks associated with Meta's 'Muse' AI agent, citing reports from Hunterbrook and Futuresociety.ch. He warns of vulnerabilities allowing users to profile third parties and reverse-engineer AI systems, predicting more negative headlines regarding data handling.

  6. 26:45 – 32:09

    Open Source Risks and Data Center Opposition

    Discussions cover the risks of open-source AI models and autonomous agents acting on user machines. The conversation shifts to public opposition against data centers, noting that resistance is often driven by local infrastructure and environmental concerns rather than opposition to AI itself.

  7. 32:09 – 32:41

    Physical Salience of Data Centers

    Speaker 3 cites research from Evan Sellinger and Divya Ramgey at the Rochester Institute of Technology, arguing that data center opposition stems from their physical scale and dystopian appearance, which allows communities to project broader anxieties onto them.

  8. 32:41 – 33:03

    Meta's Louisiana Data Center and Public Sentiment

    Elizabeth Spiers references a Meta data center project in Louisiana described as being the size of Manhattan, noting its intimidating scale. Speaker 3 confirms this and mentions other massive projects underway across the US.

  9. 33:03 – 33:59

    Rise of Tech Politics and Democratic Influence

    Speaker 3 expresses optimism that communities are connecting concerns about data centers to issues like government surveillance, economic fairness, and environmental pollution, leading to increased democratic influence over tech infrastructure.

  10. 33:59 – 34:42

    Tech Policy as a Kitchen Table Issue

    Speaker 3 notes that while professional tech policy communities have long advocated for regulation, tech policy has historically not been a decisive election issue, but is now becoming a salient 'kitchen table' issue.

  11. 34:42 – 36:01

    China AI Discourse and Safety Research

    Speaker 3 highlights interest in China's AI race and safety plans, citing publications from The Wire China, Stanford's Digichina project, and Johns Hopkins' Institute for America, China, and the Future of Global Affairs.

  12. 36:01 – 36:41

    Recursive Self-Improvement and AI Safety Alarms

    Speaker 3 discusses a Wall Street Journal piece on recursive self-improvement, noting a new paper by experts including Jeffrey Hinton and Yoshua Bengio raising alarms about AI systems automating their own research.

  13. 36:41 – 37:30

    Anthropic's Automation Numbers and Language Semantics

    Speaker 3 mentions Anthropic has released numbers regarding the automation of AI research processes. Elizabeth Spiers and Speaker 3 debate the term 'improvement,' suggesting 'mutation' may be more accurate if changes are unanticipated.

  14. 37:30 – 41:35

    AGI Claims and Timeline Predictions

    Speaker 3 suggests that before claiming AGI, labs will likely announce they have substantially automated research, leading to rapid model iteration. He references OpenAI's past AGI claim by Greg Brockman as a precedent for such announcements.

Assets discussed

AssetKindStanceContext
OpenAIprivate companyneutralWhat's revenue growth look like at Anthropic right now? What does it look like at OpenAI? Do any of these potential valuations of these IPOs make sense?
Anthropicprivate companyneutralWhat's revenue growth look like at Anthropic right now? What does it look like at OpenAI? Do any of these potential valuations of these IPOs make sense?

“Stance” is how the speakers talked about the asset, not a recommendation.

Claims worth verifying

Justin Hendrix is the CEO and editor of Tech Policy Press, an independent nonprofit publisher.

Unidentified speaker (UNKNOWN) ·1:58

How to verify: Check the 'About' page or team page on techpolicy.press for staff listings and organizational structure. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

OpenAI delayed the release of its next advanced model due to cybersecurity safety concerns.

Unidentified speaker (speaker_3) ·5:33

How to verify: Search for official announcements or press releases from OpenAI dated shortly before the podcast release (Nov 2026) regarding model delays. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Meta's Muse product allows users to build lists of people in vulnerable groups, such as teachers and military families.

Unidentified speaker (UNKNOWN) ·22:26

How to verify: Locate and read the Hunterbrook report titled 'Docs for Me Old Muse' or similar coverage in major news outlets discussing Meta's Muse capabilities. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

The US lacks comprehensive privacy legislation, relying on a patchwork of state laws.

Unidentified speaker (speaker_3) ·10:26

How to verify: Review the current status of federal privacy bills in the US Congress and compare with existing state-level privacy laws (e.g., CCPA, CPA) via government legislative trackers. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Arvind Narayanan at Princeton argues that stopping data center projects does not necessarily slow down AI advancement.

Unidentified speaker (UNKNOWN) ·30:50

How to verify: Search for publications, interviews, or social media posts by Arvind Narayanan discussing the relationship between data center infrastructure and AI progress. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

President Trump has an obsession with the idea of dominance, a trait noted by fact-checker Daniel Dale.

Unidentified speaker (speaker_3) ·16:20

How to verify: Review articles by Daniel Dale (CNN) analyzing Trump's rhetoric and behavior, specifically those discussing themes of dominance. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

The Trump administration has characterized anti-data center sentiment as a Chinese psyop.

Unidentified speaker (speaker_3) ·30:17

How to verify: Search for official statements or speeches by Trump administration officials linking opposition to data centers to Chinese influence operations. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Meta's Ray-Ban glasses were used to record people without consent and harass people upon release.

Elizabeth Spiers ·22:08

How to verify: Search for news reports from early 2026 (or relevant release period) detailing user complaints or incidents involving Meta's Ray-Ban smart glasses. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

The 'Obernulti Trehan bill' and 'Great American AI Act' are legislative proposals on the table in Congress.

Unidentified speaker (speaker_3) ·11:59

How to verify: Search the US Congress legislative database (congress.gov) for bills with these specific names or similar titles related to AI. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Futuresociety.ch claims to have pulled a copy of the virtual machine used by Meta's AI agents for reverse engineering.

Unidentified speaker (speaker_3) ·23:30

How to verify: Visit futuresociety.ch or search for reports from that domain regarding Meta's AI infrastructure and reverse engineering efforts. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Evan Sellinger and Divya Ramgey are at the Rochester Institute of Technology and argue that data center opposition is due to the physical salience of these structures.

Unidentified speaker (speaker_3) ·32:09

How to verify: Search for publications or press releases from the Rochester Institute of Technology authored by Evan Sellinger and Divya Ramgey discussing data center opposition. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Meta has a data center project in Louisiana that is the size of Manhattan.

Unidentified speaker (UNKNOWN) ·32:41

How to verify: Search for Meta's official announcements or news reports regarding their Louisiana data center project and its reported size. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Tech policy has never been a kitchen table issue or a decisive issue in most elections.

Unidentified speaker (speaker_3) ·33:59

How to verify: Review historical election polling data and campaign platforms to assess the frequency of tech policy as a primary voter concern. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

The Wire China is a publication with interesting content on China and AI.

Unidentified speaker (speaker_3) ·34:42

How to verify: Visit thewirechina.com and search for articles related to AI and China policy. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Stanford's Digichina project produces research on China and AI.

Unidentified speaker (speaker_3) ·35:02

How to verify: Visit the Digichina project website at stanford.edu and review their published research and faculty profiles. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Johns Hopkins Institute for America, China, and the Future of Global Affairs works on China-related global affairs.

Unidentified speaker (speaker_3) ·35:22

How to verify: Visit the Johns Hopkins School of International Studies website and locate the Institute for America, China, and the Future of Global Affairs. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

A Wall Street Journal piece discussed recursive self-improvement in AI.

Unidentified speaker (speaker_3) ·36:01

How to verify: Search the Wall Street Journal archive for articles discussing 'recursive self-improvement' in the context of AI systems. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Jeffrey Hinton and Yoshua Bengio are authors of a new paper raising alarms about AI systems improving themselves.

Unidentified speaker (speaker_3) ·36:21

How to verify: Search academic databases or news archives for recent papers authored by Jeffrey Hinton and Yoshua Bengio on AI self-improvement or recursive self-improvement. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Anthropic has released numbers regarding the automation of AI research.

Unidentified speaker (speaker_3) ·36:41

How to verify: Search for official announcements or blog posts from Anthropic detailing metrics on automated research processes. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Greg Brockman made a pronouncement that OpenAI had reached AGI.

Unidentified speaker (speaker_3) ·37:30

How to verify: Search for public statements or blog posts by Greg Brockman regarding OpenAI's AGI milestones. For recent events, verify 2026-08-10 through 2026-11-10. Use another period only when the speaker explicitly names it.

Do your own research

  • Verify the text and impact of the 'Obernulti Trehan bill' and 'Great American AI Act' via US Congressional legislative databases.
  • Check Hunterbrook and Futuresociety.ch reports for specific details on Meta's 'Muse' AI agent vulnerabilities.
  • Review Tech Policy Press essays by Ambersinga on 'authorization failure' frameworks for AI liability.
  • Analyze US state-level privacy laws to assess the 'patchwork' claim regarding AI privacy infractions.
  • Examine OpenAI's official announcements regarding model delays and cybersecurity safety protocols.
  • Research data center projects in Louisiana and other US states for physical scale and environmental impact reports.
  • Review publications from Stanford's Digichina project and Johns Hopkins' Institute for America, China, and the Future of Global Affairs on China's AI strategy.
  • Check Wall Street Journal articles and recent papers by Jeffrey Hinton and Yoshua Bengio on recursive self-improvement and AI automation metrics.
Full transcriptexpand
0:00
Unidentified speaker (speaker_0): BetMGM, an official sports betting partner of the National Hockey League, is ready for puck drop. Now hockey fans in Alberta can get in on the action from the opening face-off with goal in first 10, a betting market where you can call whether a goal will be scored by any team in the first 10 minutes of a game. From pregame to live bets, same game parlays, player props, and more, BetMGM has
0:20
Unidentified speaker (speaker_0): over 500 NHL betting markets to choose from. And the BetMGM app has instant deposit and fast withdrawal options, including Interact e-transfer, Visa, MasterCard and Apple Pay, and more. Download the BetMGM app and don't miss a second of the opening minutes. BetMGM, an official sports betting partner of the National Hockey League, your home for hockey
0:39
Unidentified speaker (speaker_0): bets all season long. See BetMGM.com for terms. 18 plus only, Bet Responsibly. Alberta only, subject to eligibility requirements.
0:46
Unidentified speaker (speaker_1): If gambling is affecting your mental health or well-being, 211-Alberta 211 Alberta is here to help. Call or text 211 or visit ab.211.ca. Standard message and data rates may apply. BetMGM operates pursuant to an operating agreement with the Alberta iGaming Corporation.
1:01
Unidentified speaker (speaker_0): You work hard for your business. Let your credit card work harder for you. With the WestJet RBC World Elite MasterCard for Business, unlock up to 100,000 WestJet points. Enjoy silver status for at least a year with benefits like priority boarding, lounge access, and free checked bags. Plus, earn up to six points per dollar on WestJet
1:21
Unidentified speaker (speaker_0): travel. You'll also earn up to three points per dollar on eligible business purchases, making this the fastest way to earn WestJet points. Turn business spending into your next adventure. Terms and conditions apply. Visit rbc.com/slash At
1:34
Unidentified speaker (speaker_0): WestJet Business to learn
1:36
Unidentified speaker (UNKNOWN): more.
1:47
Elizabeth Spiers: Hello and welcome to Money Talks. I'm your host, Elizabeth Spires, and today I'm joined by my friend Justin Hendricks, and we're here to talk about AI risk and policy. So, Justin, introduce yourself.
1:58
Unidentified speaker (UNKNOWN): My
1:58
Unidentified speaker (speaker_3): name is Justin Hendrix. I am CEO and editor of a site called Tech. Techpolicy.press. It is an independent, nonprofit publisher that focuses on issues at the intersection of tech and democracy. We have a small team and then a kind of set of fellows that are spread out around the world. And
2:18
Unidentified speaker (speaker_3): literally at this point, thousands of volunteer contributors, experts, folks from the field who focus on various tech accountability. issues, everyone from sort of folks in think tanks and other civil society organizations on through to academics, occasionally folks writing from industry or government, people commenting on all manner of tech
2:38
Unidentified speaker (speaker_3): policy matters.
2:39
Elizabeth Spiers: So for our listeners, I wanted to bring Justin on because of the people I know who both understand the dynamics of AI and policy and regulation, I think you sit at the nexus of all that stuff. And since we are now entering the uh panic phase that's the flip side of the hype cycle about AI, I was really interested in hearing your thoughts about
2:59
Elizabeth Spiers: what's going on. But first, I have to ask you did you see the Jaden Wickline impersonation of Dario Amade on SNL recently?
3:08
Unidentified speaker (speaker_3): Nailed it. Is the only
3:09
Unidentified speaker (UNKNOWN): thing I can say about
3:10
Unidentified speaker (speaker_3): it. Nailed it, right?
3:11
Elizabeth Spiers: So for any listeners who did not see it, comedian J. Jane Wickline showed up on weekend update as Dario Amade, and she impersonated all of his sort of, you know, interpersonal mannerisms that are a little awkward. But she started it out with: I may have been alarming in recent interviews, but I want to assure you that if
3:31
Elizabeth Spiers: we can pressure lawmakers to create guardrails, we will be able to stop me. And this strikes me as the essential conflict. that we're dealing with right now. You have all these AI CEOs who keep insisting that, yes, we do have safety concerns, but this is somebody else's problem to deal with. So to what extent is what
3:51
Elizabeth Spiers: they're saying somewhat legitimate? What how much of this is out of their hands and how much of it is it?
3:57
Unidentified speaker (speaker_3): Well, I mean, you know, listen, there are incentives and there are incentives and there are incentives, and I suppose everybody's operating. on some version of some incentive structure that is driving them forward in whatever way. But yeah, I mean, fundamentally, I think one of the reasons I kind of do what I do is I I think people have free will and agency and get to make decisions
4:17
Unidentified speaker (speaker_3): in their lives about what they choose to do, what they choose not to do. I have always thought that if you are the CEO of a company, that means that you get to make decisions about how
4:27
Unidentified speaker (speaker_3): what products,
4:28
Unidentified speaker (UNKNOWN): for instance. Yes. That's how I think we all
4:30
Elizabeth Spiers: understand it. That's
4:32
Unidentified speaker (speaker_3): and so to me that that, you know, Dario writes nice essays on his blog about pacing the frontier and sort of trying to set some constraints. But at the end of the day, I mean the guy could wake up tomorrow and decide to go slower.
4:46
Elizabeth Spiers: Yes, that's my feeling too. When I think about their incentives though. I think it's a it strikes me as a classic kind of misaligned incentives problem. You know, you have on the one hand, they have a lot of investment money. They're funding data center build-outs through debt, often using vehicles that are off balance sheet. They have to pay a lot of
5:06
Elizabeth Spiers: money in interest rates, which for the hyperscalers is less of a problem, but for the smaller AI companies, it could be. And, you know, they're they're competing with each other, which in the tech world means that whoever moves fast and possibly breaks a lot of things in route might be the ultimate winner here. And they're under pressure to compete with China. So these are all incentives that
5:26
Elizabeth Spiers: just cut against safety. Where do you see some areas where they're showing some restraint if they are?
5:33
Unidentified speaker (speaker_3): Yeah, I'm not sure we can say we're seeing much restraint at the moment. I mean, uh uh I suppose I'm talking to you a few hours or a day after OpenAI has at least said, you know, it's going to not release its next advanced model due to cybersecurity safety concerns. So you know, maybe we've learned something in the last couple of weeks, and maybe there will be a bit of a slowdown.
5:53
Unidentified speaker (speaker_3): But something tells me that delay will not be long. And you know, you look at just even some of the recent product releases. I just read a couple of reports today. Again, these are preliminary. but about Meta's Muse product and people already beginning to spot severe safety concerns around this product, which of course has come from the original Move
6:13
Unidentified speaker (speaker_3): Fast and Break Things company. So I don't know. I I feel like there's there's not a lot of evidence for these companies taking safety extraordinarily seriously, despite all the lip service that's paid to it. We have now months of headlines which would suggest that these products have been rolled out without proper
6:33
Unidentified speaker (speaker_3): considerations and that they have created a lot of potential liabilities for people.
6:38
Elizabeth Spiers: Yeah. You there's a great piece on your tech policy press site by Ambersinga.
6:43
Unidentified speaker (speaker_3): Umbersinha, yep.
6:44
Elizabeth Spiers: Called Artificial or Superintelligent, super intelligent, The Broken Metaphors of AI. And the piece argues that part of the problem here. is that all the language we use to talk about AI to lay people and to the general public, we're always using metaphors for human consciousness. And whenever we talk about quote rogue AI agents, that that's a good example
7:04
Elizabeth Spiers: because it makes it seem like the agents have agency and they don't. And so uh there's a great line from the essay that says something like a more straightforward explanation is to view the loss of control as an authorization failure. A model executing unapproved commands is not demonstrating rogue agency. It's experiencing a permission architecture failure.
7:24
Elizabeth Spiers: And I thought that was so good because it puts the responsibility back on the the CEOs and the people who are creating this architecture. So how do we how do we f restraint restrain I don't want to use the force, but like how do we force them to actually do the things that they need to? tighten that permission architecture?
7:41
Unidentified speaker (speaker_3): Well, I think you know, you've got a couple different ways that you could potentially go about that. You can, of course, introduce new laws, or you could potentially wait for liability to catch up with these companies. You know, right now it's sort of unclear exactly who will face the first significant lawsuit about one of these agents
8:01
Unidentified speaker (speaker_3): operating and creating a vulnerability for a corporation or a government or some individual. And there are some significant questions, I think, when you kind of dig into the law about exactly how liability will work when it comes to AI agents. And it's kind of the layers of potential responsibility between, you know,
8:20
Unidentified speaker (speaker_3): users who may be deploying those agents, the companies who are developing them, et cetera. But it feels to me like those. are the two kind of methods we've got here. You can have some new laws, or you can wait for the the legal system to kind of hash it out and to to kind of figure it out. Look at where we've got to with social media. Lots of folks kind of spotted
8:40
Unidentified speaker (speaker_3): various problems with social media products and platforms over the years. And while we're seeing now a spate of successes among those who have taken those companies to court and tried to address those types of issues, how long has it been? I mean, it's been, you know, a really, really long time. I suspect that we don't have quite
9:00
Unidentified speaker (speaker_3): that long for things to work out through the courts with AI. So we'll see. But I think another big question is: when will something happen that is so severe or so significant that it creates motivation? either among lawmakers or in the companies themselves to take more swift
9:20
Unidentified speaker (speaker_3): action?
9:20
Unidentified speaker (UNKNOWN): Yeah,
9:21
Elizabeth Spiers: that's what uh you know disturbs me is the tech industry in particular is notorious for waiting for something bad to happen before they do anything to shore up safety concerns. And I think of Meta is as the sort of pioneer of doing that and then being completely unapologetic about it. So we have that problem with just getting the tech
9:40
Elizabeth Spiers: CEOs to look proactively at things that might be problems and ideally preempt them, but that's not how they typically operate. But also with legislation, you know, we often are reactive and we have a uh Congress that is not super tech literate to begin with. So what's the starting point on the policy side? What do we need to do?
10:00
Elizabeth Spiers: Is it educating lawmakers? Is it building public
10:03
Elizabeth Spiers: outrage enough that they feel like they have to act? Aaron
10:06
Unidentified speaker (UNKNOWN): Powell
10:06
Unidentified speaker (speaker_3): I don't have a a specific legislative solution or specific set of priorities that I would personally advocate, but I do think that a lot of the pieces are still on the table or from the past that would be necessary to kind of implement. And there have been a lot of innovative legislative proposals that have been put forward lately
10:26
Unidentified speaker (speaker_3): that at least deserve your listeners' kind of consideration. But in the US, one of the things to just remember is that some of the absolute basics that we would need to have in place to prevent the worst harms from artificial intelligence are simply not in place. We do not have comprehensive privacy legislation in this country. And you might argue, oh, you know,
10:46
Unidentified speaker (speaker_3): many of the states have introduced their own privacy laws and they've kind of filled in the gaps somewhat. I don't think most people who look at these questions would agree that the state patchwork has effectively immunized us to the worst potential results of privacy infraction related to artificial intelligence. So all this the information, this personal
11:06
Unidentified speaker (speaker_3): information that's sloshing around all over the place, um, that's a major problem. And there are various other kind kinds of basics that we simply just don't have in place here. But even in a place like your you know, in Europe, where they've put so much effort into trying to have a comprehensive suite of legislation that lets them effectively govern technology, they don't feel they're quite
11:26
Unidentified speaker (speaker_3): ready. You know, there's a lot of lawmakers there now looking at AI agents and saying, is the AI Act, which we just passed, is that Does fit? Does that effectively give us the liability tools that we need? So I don't think anyone's anyone
11:38
Unidentified speaker (UNKNOWN): not a good idea.
11:39
Unidentified speaker (speaker_3): has quite got this figured out, but we need to go back to basics. And then I think we do need to look at some of these more serious efforts, things like the Obernulti Trehan bill, the Great American AI Act. The Trehan Obernulti bill does at least represent a kind of effort, a swing at a comprehensive suite of
11:59
Unidentified speaker (speaker_3): legislation. It's at least an attempt. And people will, I think, will have to pick apart the different parts of it and certainly try to come to some conclusion about which way is the best way forward on some of the details. But there are at least legislative proposals like that that are that have come in front of Congress that are on the table. Aaron
12:15
Unidentified speaker (UNKNOWN): Powell So
12:16
Elizabeth Spiers: what are what are some of the proposals that either misunderstand the problem or just oversimplify it?
12:22
Unidentified speaker (UNKNOWN): Well,
12:22
Unidentified speaker (speaker_3): I mean I think one you're seeing multiple instances of these kind of kills. kill switch type bills that are coming along. And a lot of folks who look at these things closely feel that they more or less miss miss the target, partly because they may not be technically feasible, but also partly because, you know, in a way they kind of adopt this notion
12:42
Unidentified speaker (speaker_3): of a sort of killer AI or existential risk from artificial intelligence and imagine that for whatever reason such risks come to be that pulling the plug on some level will be the answer. And that does seem like a fantasy, right?
12:56
Elizabeth Spiers: Yeah. It also just seems like something that would be impossible to do on a common sense basis.
13:01
Unidentified speaker (speaker_3): That's right. And I don't exactly know where we've got to with how every artificial intelligence model is managed or you know what the data center architecture is for OpenAI or an anthropic, but the idea that you know under Dario's desk there'll be a switch that, you know, if things get bad, he can just simply press the button. I suspect it's not that what that way.
13:21
Elizabeth Spiers: Yeah. One of one of the things that irritates me about the debate is that all these incidents where the agents get out of the sandbox are all instances where this isn't an issue of an AI company rolling out a model that already has these problems to the public. These are things that are happening in the frontier labs in supposedly secure environments because they are
13:41
Elizabeth Spiers: willing to turn up the temperature of the AI to 11 and just see what happens. And to me that again just points to corporate responsibility. But when the CEOs are saying, okay, well, we're waiting to be regulated, there's just not much to stop them there. So what what's your personal p doom in the current environment?
14:01
Unidentified speaker (speaker_3): It's almost like one of those questions that I find myself just not able to even consider on some level. I think because I'm so hung up on the idea that I don't need to be afraid of artificial intelligence. I need to be afraid of men with very irresponsible approaches to
14:20
Unidentified speaker (UNKNOWN): the
14:20
Unidentified speaker (speaker_3): deployment of artificial intelligence. That is much more my concern. And that those men are in Silicon Valley, but they're also in the Pentagon. And maybe they're also in Beijing. I don't know precisely where they all are. But I'm chiefly concerned about people behaving badly with AI than I am about AI
14:40
Unidentified speaker (speaker_3): itself. Aaron
14:41
Unidentified speaker (UNKNOWN): Powell
14:41
Elizabeth Spiers: Yeah.
14:45
Elizabeth Spiers: Analogous in terms of the way that they think about AI. And I think Trump's recent obsession with it is just that he he's finally internalized the idea that the stock market highs are heavily dependent on AI spending.
14:56
Unidentified speaker (speaker_3): I think you might be right about that. He very much is a booster of the technology. And we could speculate as to why it might be because he realizes the entire economy hinges on AI continuing to carry on attracting the level of investment and the level of infrastructure development that. it has. It could be
15:16
Unidentified speaker (speaker_3): because his family is apparently heavily invested in various AI firms and AI in related interests. It could be because many people that in his cabinet appear to have various interests that certainly intersect with AI. It could be because of all the kind of political spending that has gone
15:36
Unidentified speaker (speaker_3): in his favor from AI. AI-oriented interests in Silicon Valley. Or it could be really that he is truly excited about what he wants
15:44
Unidentified speaker (UNKNOWN): to call
15:45
Unidentified speaker (speaker_3): superintelligence. I don't know.
15:46
Elizabeth Spiers: I I
15:47
Unidentified speaker (UNKNOWN): can't
15:47
Elizabeth Spiers: imagine that a guy who famously does not use a computer and kind of shoes technology and has someone else post to his social media has somehow developed a lot of affection for chatbots in the lowest level version of AI.
16:00
Unidentified speaker (speaker_3): I think you're probably right. I mean, I think if you look you look at Donald Trump and you kind of think about like what seems to drive the guy, he's I was reading something by uh Daniel Dale, you know, the fact checker who's now, I believe, at CNN, who's been following Trump for 15 years and kind of trying to tally all of his various lies. And Daniel Dale was pointing
16:20
Unidentified speaker (speaker_3): out that But one of the things that's consistent about many of the lies he tells, especially those ones that start off with. someone addressing him as sir, or with a burly man who breaks down in tears. That's like
16:31
Unidentified speaker (UNKNOWN): a kind of
16:32
Unidentified speaker (speaker_3): constant trope. Trump has an obsession with like the idea of dominance. And that those types of stories are like always bringing that to the fore, the idea of dominance. And you know, s AI in many ways, like it also has this kind of like aroma of dominance around it. Like if we we do it really right, we'll be the most dominant character. which is
16:51
Unidentified speaker (speaker_3): the word that they use in their AI plan, their AI policy. We're pursuing unquestioned dominance.
16:59
Elizabeth Spiers: Well, this is another situation where language really defines the way that people are thinking about it. It's really driving the framing. So when you talk about the race to dominance, you know, it implies that there's one finish line and that there's one goal. How do you think about it? Is there is there a use? useful way to frame it so that people don't understand it as a zero-sum game,
17:19
Elizabeth Spiers: which is another Trump thing. He only understands anything it through that lens.
17:23
Unidentified speaker (speaker_3): Yeah, listen, I have colleagues and of course contributors to Tech Policy Press that are smarter on the technology and can probably tell you in more technical detail precisely why we shouldn't be expecting super intelligent. artificial intelligence in the near term. And they'll probably start by
17:43
Unidentified speaker (speaker_3): questioning whether that is even an idea. It's even something that we should want to pursue, or that it could exist, or whether transformer models could never get to that level of of human intelligence to begin with, et cetera, et cetera. We can get into all that detail on some level. But I don't know. I just I don't imagine a world. where
18:03
Unidentified speaker (speaker_3): suddenly you wake up and you've got an AI that has just completely obliterated the idea of competition, you know, uh between powers or has completely neutralized the reality of the physical world that we live in, the distance between things, the difficulty of like, I don't know,
18:22
Unidentified speaker (speaker_3): intervening in the physical world. I just don't see that, at least in the near term, right? Me
18:27
Elizabeth Spiers: neither. Well, you know, you and I met, I think when we were both working on VR projects. And this feels like a cycle. Anytime some new technology comes out, the the people with corporate interest stand up and say, oh, this is great for society. And and they always make exactly the same promises. Like you're never going to have to work. We'll do UBI. We've seen this so many times before.
18:47
Elizabeth Spiers: I think we saw it I the first time I remember it in my professional life was probably the beginning of the dot com boom.
18:54
Unidentified speaker (speaker_3): Yeah, I don't exactly know whether we're in the biggest bubble ever right now, or whether we're in some kind of, you know, maybe more late 20th century dot com style bubble, which will burst and cause a lot of pain, but still leave a bunch of really interesting companies and technological innovations and
19:14
Unidentified speaker (speaker_3): fundamentally research. reshape the economy forever and we'll continue to see all of manner of growth and investment as a result. I don't know exactly what's going to happen with artificial intelligence. I don't think really anybody does. I see a lot of people really trying to game it out.
19:30
Unidentified speaker (speaker_3): What's revenue growth look like at Anthropic right now? What does it look like at OpenAI? Do any of these potential valuations of these IPOs make sense? Does any of this data center investment make sense? You know, I think if anyone had figured all that out, they certainly wouldn't be taking the subway to work like I do every day. But I don't know. I just don't
19:50
Unidentified speaker (speaker_3): know what to anticipate. So I don't know what that future looks like. I know what I'm worried about. So
19:55
Unidentified speaker (UNKNOWN): what
19:55
Elizabeth Spiers: are what are your biggest worries?
19:57
Unidentified speaker (speaker_3): It again, it's concentration of power. It is effectively a world where you've got a handful of companies with extraordinary power and leverage over the economy and ultimately leverage even over the United States government. It's a world where, you know, that kind of distinction between the U.S. government and its major
20:16
Unidentified speaker (speaker_3): technology firms is difficult to make. You know, a world where every other power on the planet passed it aside, is it with us or against us? Those types of things, I think. are worth being afraid of. Aaron
20:28
Unidentified speaker (UNKNOWN): Powell
20:28
Elizabeth Spiers: What uh what do you think consumers need to think about in terms of data privacy and AI? Aaron
20:32
Unidentified speaker (UNKNOWN): Powell
20:33
Unidentified speaker (speaker_3): Again, there are smarter people than me to make these types of recommendations, but there are some obvious ones, you know. I mean, I I read that review of of Meta's Muse in the New York Times a few days ago where the journalists had tried it out, and I think was it reasonably open about the idea that it was a huge privacy risk. And I think talked to a couple of people who were like, Nope,
20:52
Unidentified speaker (speaker_3): we'll never try that, not handing over that kind of information. But you know, these tools are extraordinary gobblers of personal information and data. And so if you install one of these things onto your computer and you give it permissions and then you walk away, I think you should anticipate something bad happening at some point. You may not know what it is,
21:12
Unidentified speaker (speaker_3): right? But the idea that this company again, slowed down and put in every type of consideration necessary before getting this product out into the marketplace in in short order in this kind of competitive environment, I think that's likely untrue. And so I would actually, if there's any one little prediction I'll make here, it's that
21:32
Unidentified speaker (speaker_3): we'll see a spate of headlines over the next few weeks or months about, if not Muse, products like Muse that are increasingly concerning about what they're doing either with their users' personal information or with the personal information of others.
21:48
Elizabeth Spiers: Yeah, I have a theory that whenever people are developing technology, especially big companies like Meta, then and they're writing user stories to kind of figure out the architecture of the product, they almost never have a user story that says, I'm a giant asshole and I want to do as much damage as possible. I mean, there are so many bad actor possibilities for technologies
22:08
Elizabeth Spiers: like these. And we saw this when Meta rolled out their Ray Ban glasses, the meta glasses. Immediately people started using it to r record people without their consent and harass people. This is a little off topic for AI, but what were some of the well, no, it's not. What were some of the the things that you read that concerned you about me Muse? as
22:26
Unidentified speaker (UNKNOWN): soon?
22:26
Unidentified speaker (speaker_3): So, and by the way, are you familiar with Hunterbrook?
22:29
Elizabeth Spiers: Yes.
22:30
Unidentified speaker (speaker_3): Okay. So on uh the Hunterbrook website today they have a piece called Docs for Me Old Muse. Meta's new AI agent built lists of people in vulnerable groups on request. And so this is an account effectively of how the users Us effectively were able to go and profile third parties, build
22:50
Unidentified speaker (speaker_3): lists, of Facebook and Instagram accounts across different communities, everybody from teachers to pro-Palestinian activists to military families, et cetera, et cetera. And just the extent to which the agent was totally cool with meeting these types of user prompts. That is concerning. Again,
23:10
Unidentified speaker (speaker_3): I haven't personally reviewed all of this material that the folks at Hunter Hunterbrook B. are reporting on, but it is something. I'll say that there's another thing that I looked at today, and again, not entirely aware of exactly how this was done, but another group operating at a website called Futuresociety.ch,
23:30
Unidentified speaker (speaker_3): which claims it's kind of pulled down a copy of the the virtual machine that Meta's AI agents using and essentially tried to reverse engineer some of it. And some of the kind of considerations around privacy and and the way the thing is is sharing or otherwise not appropriately kind of
23:50
Unidentified speaker (speaker_3): sandboxing information is really concerning. So I don't know precisely like where we'll get to on this, but something tells me that those types of security considerations we'll will just consit continue to see more and more exploration of it and more and more bad headlines.
24:27
Unidentified speaker (speaker_1): Legendary
24:29
Unidentified speaker (speaker_0): has arrived in Alberta
24:30
Unidentified speaker (UNKNOWN): with BetMGM
24:31
Unidentified speaker (speaker_0): Casino.
24:32
Unidentified speaker (UNKNOWN): Sign up to
24:33
Unidentified speaker (speaker_0): unlock access to over 4,000 casino games, including exclusive slots you can't spin
24:38
Unidentified speaker (UNKNOWN): anywhere else.
24:39
Unidentified speaker (speaker_0): Survivors
24:40
Unidentified speaker (UNKNOWN): ready?
24:40
Unidentified speaker (speaker_0): The
24:40
Unidentified speaker (UNKNOWN): iconic TV show is now an online
24:43
Unidentified speaker (speaker_0): casino
24:43
Unidentified speaker (UNKNOWN): game where you
24:44
Unidentified speaker (speaker_0): can tap into island To
24:45
Unidentified speaker (UNKNOWN): fun.
24:45
Unidentified speaker (speaker_0): BetMGM
24:46
Unidentified speaker (UNKNOWN): has
24:47
Unidentified speaker (speaker_0): also brought the prices
24:48
Unidentified speaker (UNKNOWN): right:
24:48
Unidentified speaker (speaker_0): Game of Thrones, Family Feud, and MGM Millions, and so many more exclusive casino games to Alberta. Get Las Vegas right from your phone with live dealer games streaming from real tables. And every time you play, you can enjoy instant deposit options and fast withdrawals. BetMGM Casino and its legendary library of games are
25:08
Unidentified speaker (speaker_0): available. Download the BetMGM app. See BetMGM.com for terms. 18 plus Alberta only. Please play responsibly, subject to eligibility requirements.
25:19
Unidentified speaker (speaker_1): If gambling is affecting your mental health or wellbeing, 211 Alberta is here to help. Call or text 211 or visit ab.211.ca. BetMGM operates pursuant to an operating agreement with the Alberta IGaming Corporation. Standard message and data rates may apply.
25:33
Unidentified speaker (speaker_0): You work hard for your business. Let your credit card work harder for you. With the WestJet RBC. World Elite Mastercard for Business, unlock up to 100,000 WestJet points. Enjoy silver status for at least a year with benefits like priority boarding, lounge access, and free checked bags. Plus, earn up to six points per dollar on WestJet
25:53
Unidentified speaker (speaker_0): travel. You'll also earn up to three points per dollar on eligible business purchases, making this the fastest way to earn WestJet points. Turn business spending into your next adventure. Terms and conditions apply. Visit rbc.com/slash WestJet Business to learn more. Optimize Wealth Management was built for ambitious investors reaching for the next level
26:13
Unidentified speaker (speaker_0): of their financial future. Optimize leverages sophisticated proprietary investment strategies and provides its clients with next generation financial planning, delivered by a team focused on meaningful results, not sales pitches. Visit OW. OWM.com slash clients to learn more. Optimize wealth management for those who want more.
26:45
Unidentified speaker (speaker_3): Claude. These days they would they do spin up agents if you're doing code projects or things of that nature. They do spin up sub-agents to do various types of tasks. I have not used any of the types of agents like Muse or Hermes or OpenClaw or any of the
27:01
Unidentified speaker (UNKNOWN): other
27:01
Unidentified speaker (speaker_3): types of things that effectively take over your machine and you know you grant access to all manner of permissions across your various suite of software.
27:11
Elizabeth Spiers: Yeah. I did a story back in February about an engineer who who was monitoring some kind of tech library, I forget what. But another person had spun up an agent to submit code to it. And then when the engineer rejected the agent's code, the agent went into the system on GitHub and started writing a hit piece about the engineer and posted
27:31
Elizabeth Spiers: repeatedly. And yeah, the engineer thought it was funny, but it took a while to clean up and also figure out who was controlling the agent. And he was sort of just stunned that people would give an agent that much license or control. So this strikes me as a different problem than, you know, what are these CEOs going to do, especially if we're talking about open
27:50
Elizabeth Spiers: source software or things that are not closed systems, not limited to begin with? How do you think about risk in that context?
27:58
Unidentified speaker (speaker_3): I think we probably just don't even understand what's going on or what's happened. I mean, it's it just feels to me like we're in a tip of the iceberg situation. These types of disclosures from OpenAI, from anthropic, from various researchers and labs who are following these things closely,
28:12
Unidentified speaker (UNKNOWN): you know,
28:12
Unidentified speaker (speaker_3): this stuff in it's voluntary, except in the say situations perhaps like Hugging Face, where you know, in some way it came out before OpenAI AI even realized its involvement effectively. But in many circumstances, I reckon that things are go happening or going on that no one has noticed yet, or you know, we don't even know about.
28:33
Unidentified speaker (speaker_3): And, you know, I could looked at that Anthropic threat report, which had all those extraordinary things that have that people have used Anthropic to do across the world, you know, everybody from the Houthis on through to Iran on through to China and Russia, et cetera, et cetera. And all that stuff's headline grabbing. But, you know, many. of those things, it's important to remember, you
28:53
Unidentified speaker (speaker_3): could do those same things with an open weight model that you downloaded from Hugging Face and ran on your laptop. So it's not clear to me exactly how we would ever understand precisely what has happened just in the last few weeks or months. Aaron
29:07
Unidentified speaker (UNKNOWN): Ross Powell
29:07
Elizabeth Spiers: If there were some recommendations that you had for voters, what what should what kind of pressure should they put on Congress to recommend? regulate? What would be the best line of attack?
29:18
Unidentified speaker (speaker_3): Well, I mean, I would say that we're beginning to see what looks like a really interesting kind of political formation of people who are concerned about artificial intelligence across the board, concerned about data centers, concerned about surveillance, concerned about AI in schools, concerned about AI in the
29:38
Unidentified speaker (speaker_3): workforce. and in the workplace. So maybe one step back from what would I tell them they should say to Congress, I think I'm observing how that kind of political formation evolves. Where does it get to? Is it able to remain sort of, I don't know, cross partisan? Doesn't seem right to call it
29:57
Unidentified speaker (speaker_3): bipartisan or something, but
29:59
Elizabeth Spiers: Yeah. Yeah. It's also funny to me that now that we have one bipartisan issue that we can all agree on, which is that everybody hates data centers, the Trump administration's reaction to that is to basically say, well, all of this anti-AI sentiment, anti-data center sentiment is just a Chinese psyop.
30:17
Unidentified speaker (speaker_3): Yes. And we have, of course, seen various efforts to kind of paint data center opposition as the result of foreign influence.
30:26
Unidentified speaker (UNKNOWN): So
30:26
Elizabeth Spiers: data centers. have a lot of problems besides public opposition. One is just that let's say you are able to do this enormous build-out that's already being financed. In a lot of cases, these data centers are not going to even be able to get online because we don't have the infrastructure to support it. If the data centers don't get build out, does
30:46
Elizabeth Spiers: that slow down Frontier Labs just by default?
30:50
Unidentified speaker (UNKNOWN): This
30:50
Unidentified speaker (speaker_3): seems to be a point of argument. I've seen arguments like from Arvind Narayanin at Princeton that suggests that there's not much that you can do. Stopping a data center project doesn't necessarily slow down the advance of artificial intelligence all that much. But I'm not sure that's really what the goal of most anti-data
31:09
Unidentified speaker (speaker_3): center activists activist really is.
31:12
Elizabeth Spiers: Yeah, I don't think it has anything to do with that. I think it's they've heard horror stories about their electric bills going up, the the data centers are noisy, their pollution issues, all sorts of things, and they just don't want them in their neighborhood.
31:24
Unidentified speaker (speaker_3): Yeah, and they may disagree simply with the way that it's come to town. Maybe they don't
31:29
Unidentified speaker (UNKNOWN): like
31:29
Unidentified speaker (speaker_3): the fact that it was not transparent and all done by through an NDA or they don't like how the land project was acquired. Or they don't like the fact that the state decided to subsidize the hell out of the thing, or they don't like the deal that the local government did to provide it with electricity at some preferential rate or whatever it might be. There's all kinds of reasons that that
31:49
Unidentified speaker (speaker_3): people might oppose these situations that don't really have much to do with opposing artificial intelligence. I do think that there is a kind of sense among many that resisting data centers is also a way of sort of saying this is not the future. we want. Yeah. Like this is not the economy we want. And there's something about them. We had a piece a couple of weeks ago
32:09
Unidentified speaker (speaker_3): from Evan Sellinger and Divya Ramgey, both at the uh Rochester Institute of Technology, who were kind of making this argument that one of the reasons that data center opposition has kind of taken such shape is because of the literal physical salience of these things. in communities. They're so big
32:29
Unidentified speaker (speaker_3): honking boxes. They look dystopian. They look scary. And you can put on them in many ways. You can transfer onto them, you know, many of your concerns about the future in the world that we're building.
32:41
Unidentified speaker (UNKNOWN): Yeah.
32:41
Elizabeth Spiers: There's a I think Meta has one that's supposed to be the size of Manhattan. And that's just mind-boggling. That's that's that really does feel like something out of a William Gibson novel.
32:52
Unidentified speaker (speaker_3): Absolutely. I think you're talking about the uh project in Louisiana, and there are other very massive ones that are, of course, underway in different parts of the country.
33:01
Elizabeth Spiers: What are you optimistic about in this environment?
33:03
Unidentified speaker (speaker_3): Well, I I think I would look to that kind of process that we're in where people are kind of recognizing that there are some similarities between concerns. around things like government surveillance, things like am I getting a fair shake in this economy? Things like, is the data center that's being
33:23
Unidentified speaker (speaker_3): built down the street damaging the environment and spewing out lots of pollution from the gas turbines that have been set up to get it speedily underway? I think people are beginning to kind of connect those threads, connect those dots, and they're looking for ways to have a democratic
33:39
Unidentified speaker (speaker_3): influence over the future of tech. And that I think is something that, you know, for all the effort of the kind of professional tech policy community, the civil society organizations, many of the folks that we both write for and who literally write the material that we publish at Tech Policy Press all the time, they haven't really been able to
33:59
Unidentified speaker (speaker_3): manufacture. Tech policy in many ways has never been a kitchen table. issue. It's certainly never been the most salient issue in an election or a decisive issue in most elections, right? And the fact that it's become so, I think it's extraordinary. And I I do think that it suggests that we're in a kind of different moment. In a way, like uh
34:19
Unidentified speaker (speaker_3): I heard at an event I attended earlier this year, this line, which I'll probably not get quite right, but it's almost like tech policy is not so interesting, but tech politics is on fire.
34:30
Elizabeth Spiers: Yeah. So for listeners, tech policy press is a tech policy.
34:35
Elizabeth Spiers: And everyone should read it. Who are you reading right now who you think is really interesting and thoughtful about all of this?
34:42
Unidentified speaker (speaker_3): Oh gosh. I suppose I'm probably always looking to slightly uh even even nerdier places to get my information. I find myself really interested. lately in the China discourse, China and AI race, and there's a a publication called The Wire China that seems to have a bunch of interesting stuff on it. I was just reading an interesting
35:02
Unidentified speaker (speaker_3): essay uh from someone called Brian Chu on how China plans to keep AI safe. I try to pay attention to the good folks over at Stanford's Digichina project on that particular score. They always have a lot of really interesting stuff. The folks at Johns Hopkins, Institute for America, China, and the Future of Global Affairs
35:22
Unidentified speaker (speaker_3): at the School of International Studies who are working on that. So I don't know. That's one area where I'm just really trying to kind of make sure I pay attention to experts and try to really understand. I find like, I don't know, most of us Americans, we have very flat understanding of China. I've only been to that country one time. I've
35:42
Unidentified speaker (speaker_3): traveled to two cities there. I feel like I have a very minimal understanding of the many, many, many complexities of China and its politics. And I don't know. I just find I have to perpetually invest in trying to overcome that.
35:56
Elizabeth Spiers: Before we wrap up, is there anything that you'd like to talk about that we didn't discuss?
36:01
Unidentified speaker (speaker_3): Oh, that's a good question always. Well, I'll say one thing. We're always kind of like looking at what are the labs talking about and what seem to be some of the kind of questions that people are trying to answer. And there was a piece in the Wall Street Journal about this kind of question around how AI systems improve themselves.
36:21
Unidentified speaker (speaker_3): So you often hear this phrase recursive self-improvement, and there's a new paper out from a bunch. of different experts, including some of the chief researchers at the big AI labs and characters like uh Jeffrey Hinton and Yashua Bangio and others. And they're really trying to raise the alarm about the prospect that these AI systems may
36:41
Unidentified speaker (speaker_3): quote unquote improve. We might argue about what improvement really is in this context, but they improve themselves or that various aspects of AI research will get auto-surf. automated. And yeah, they're offering various numbers about, you know, how much of that process already is automated. Anthropic's thrown out some numbers. We have no idea if any of this is, you know
37:01
Unidentified speaker (UNKNOWN): This
37:02
Elizabeth Spiers: is again where language matters. You know, they're sort of trying to introduce improvement as is almost a term of art, but you could also call it mutation, especially if it goes in directions that nobody's anticipating.
37:12
Unidentified speaker (speaker_3): Absolutely. It's another one of those exactly as you say, like what is improvement and what are we aiming towards? Improvement suggests there's some future state that we've agreed is a state of being improved.
37:24
Unidentified speaker (UNKNOWN): Yeah.
37:24
Elizabeth Spiers: Or that it inherently gets better, which we sometimes it doesn't. It's not a not a given.
37:30
Unidentified speaker (speaker_3): Yeah. But on the path to AGI, I suspect someone's going to come along and say, before they even claim AGI, and I realize that open OpenAI AI just kind of sort. of claimed that they'd reached AGI, but I think most people have looked past Greg Brockman's pronouncement on that. But you know, I think you will see one or more of the labs come along at some point soon and say,
37:50
Unidentified speaker (speaker_3): we have substantially automated research. And so we are going to be in the in a situation where now things are going gonna to move extremely quickly. And that'll be a kind of prelude to one of them saying, we've achieved AGI.
38:03
Unidentified speaker (speaker_3): Or we can at least put a timeline, an exact timeline on it now. We think that we'll be able to introduce the next model and the next model and the next model at an extremely fast rate because we're ventured into this loop.
38:17
Unidentified speaker (UNKNOWN): It
38:17
Elizabeth Spiers: feels like the goalpost for what constitutes AGI keep moving. What's your definition of it?
38:23
Unidentified speaker (speaker_3): Well, I mean, I think most lay people would understand it as
38:27
Unidentified speaker (speaker_3): reaching a point where you've got some kind of software that can reasonably accomplish most of the cognitive tasks that a human adult would be able to accomplish. I think that definition is not too far off from what some of the labs have, but the reality is that But I don't know how we would ever get
38:47
Unidentified speaker (speaker_3): to the point of being able to evaluate that, nor would
38:50
Unidentified speaker (UNKNOWN): we have a value.
38:51
Unidentified speaker (speaker_3): be able to necessarily agree even on what human intelligence really consists of. So I don't know. To me, I kind of side more with the pesky nerds who suggest that even thinking about it is probably not that valuable a thing.
39:07
Unidentified speaker (UNKNOWN): All
39:07
Elizabeth Spiers: right. Well, thank you for being here.
39:11
Unidentified speaker (speaker_3): Elizabeth, thank
39:12
Unidentified speaker (UNKNOWN): you.
39:13
Unidentified speaker (UNKNOWN): And
39:13
Elizabeth Spiers: that's it for Money Talks. Thank you to our guest, Justin Hendrix, for joining, and thanks to Justin Molly of Seaplane Armada for producing. We'll be back on Saturday with our regular edition of Slate Money.
39:33
Unidentified speaker (speaker_0): BetMGM, an official sports betting partner of the National Hockey League, is ready for puck drop. Now hockey fans in Alberta can get in on the action from the opening face-off with goal in first 10, a betting market where you can call whether a goal will be scored by any team in the first 10 minutes of a game. From pregame to live bets, same-game parlays, player props, and more, BetMGM
39:53
Unidentified speaker (speaker_0): has over 500 NHL betting markets to choose from. And the BetMGM app has instant deposit and fast withdrawal options, including Interact e-transfer, Visa, MasterCard, and Apple. Pay, and more. Download the BetMGM app and don't miss a second of the opening minutes. BetMGM, an official sports betting partner of the National Hockey League, your home for hockey
40:12
Unidentified speaker (speaker_0): bets all season long. See BetMGM.com for terms. 18 plus only, bet responsibly. Alberta only, subject to eligibility requirements.
40:20
Unidentified speaker (speaker_1): If gambling is affecting your mental health or well-being, 211 Alberta is here to help. Call or text 211 or visit AB.211. ab.211.ca. Standard message and data rates may apply. BetMGM operates pursuant to an operating agreement with the Alberta iGaming Corporation. What's
40:34
Unidentified speaker (speaker_3): the
40:34
Unidentified speaker (speaker_1): best thing about the holidays? Snow days? Eggnog. How about cashback on just about everything you need to make the holidays feel? Well, holiday. With Rackiton, you can earn cash back on gifts for everyone on your list, from clothes and electrical. electronics to toys. At stores you love like Best Buy,
40:54
Unidentified speaker (speaker_1): Sephora, Walmart, and Lenovo. You save even more when you stack up to 15% cash back on top of the season's biggest sales. That's savings on savings. And beyond the gifts, you can save on decor, travel, experiences, and maybe a little something for yourself. It is the most rewarding time of the year, after all. So
41:15
Unidentified speaker (speaker_1): make your list and save on it twice. Just shop the stores you love, and cashback is automatically added to your account. And when it's time to redeem your earnings, choose from Interact, PayPal, or Check. Join for free at racketon.ca or get the Rackettin app. That's R-A-K-U-T-E-N.
This page was produced automatically: the audio was transcribed with nvidia/parakeet-tdt-0.6b-v3+sortformer-v2.1 and interpreted by Qwen3.6-35B-A3B-UD-IQ3_S.gguf. Transcription and interpretation can contain errors; always confirm against the original episode. Nothing here is investment advice.

More from Slate Money

Slate Money

The Nvidia Honeypot Boils Over

Episode discussion: NVIDIA CEO ousting, Anthropic IPO, and state tax debates dominate Slate Money

Speakers’ claims: Felix Salmon and Emily Peck discuss the resignation of NVIDIA CEO Jensen Huang amid personal scandals, noting the stock's initial drop and recovery. They analyze the corporate governance implications, contrasting Huang's lack of dual-class stock with other tech founders. The conversation shifts to the rapid pace of market narratives, highlighting Anthropic's upcoming IPO and the resilience of AI sector momentum. Finally, the hosts cover Washington State's new millionaires tax, the legal challenges of wealth taxes, and consumer alerts regarding BMW recalls.

NVIDIAJensen HuangCorporate Governance
55 min1 assets17 claims to verify
Slate Money

Political Ad Splurge

Episode discussion: 2026 Midterm Ad Spend, Texas Election Shifts, and AI Regulatory Risks

Speakers’ claims: Sarah Fischer reports that $11.87 billion was spent on political ads in the 2026 midterms, a record for a midterm cycle but lower than the previous presidential cycle. She attributes the spending surge to Citizens United and legislative deadlock, noting that local TV received $5.5 billion. Felix Salmon discusses market volatility following the election, highlighting a sell-off in core AI stocks like CoreWeave due to data center moratorium concerns. The hosts also analyze Elon Musk's failed $300 million effort to keep Texas red and the failure of California's Proposition 40 wealth tax.

Political Advertising2026 MidtermsTexas Elections
1h 4m2 assets18 claims to verify
All insights