80,000 Hours Podcast

1. Tom Milton on how we could verify a US–China deal on AI

   October 23, 20261 h 49 min

   If the US and China agreed to pause frontier AI training, how would either side confirm the other was complying? Today, Tom Milton argues, the most likely answer would be to shut down data centres entirely …

   Show notes

   If the US and China agreed to pause frontier AI training, how would either side confirm the other was complying? Today, Tom Milton argues, the most likely answer would be to shut down data centres entirely. That would throw away everything else those chips do, and make any agreement far more costly to accept.

   Tom is CEO of Amodo Design, a UK hardware lab building prototype tools to verify how AI chips are being used. He argues that verification is what makes a slowdown affordable: if a cluster can be shown to be serving only existing models rather than training new ones, most of AI’s benefits can continue during a pause.

   Amodo’s pilot reruns a cluster’s inference on a separate verifier server and checks that all of its network traffic can be explained by the agreed model. AI companies are developing related tools: in its October 2026 Frontier Safety Roadmap update, Anthropic reported a prototype that runs on production inference and checks that outputs come from a specific set of model weights.

   The field is small. Tom estimates that only about 50 engineers worldwide work on compute verification, compared with the billions of dollars the US spent on Project Vela to detect Cold War nuclear tests. And custom verification hardware has long lead times: it has to be designed, security-tested, red-teamed by both sides, manufactured, and trusted before anyone can rely on it.

   If a deal is struck in a hurry, Tom expects it to be assembled from whatever tools already exist. Were that to happen next month, he says, the only option would be to shut data centres down entirely. That is why he argues the work needs to scale up now, with more people and resources from AI companies, governments, and funders, so that better options are ready as soon as a deal is needed.

   Tom and host Tom Reed also discuss how an AI company might evade verification, lessons from nuclear and chemical arms control, and how AI is changing hardware engineering.

   Learn more about Amodo’s work at ai-verification.com.

   
   

   Links, video, and full transcript: https://80k.info/tm

   This episode was recorded on September 29, 2026.

   Chapters:

   • Cold open: Can we verify a US–China AI deal? (00:00:00)
   • Why compute is the best place to verify an AI slowdown (00:01:04)
   • Can you prove a lab isn’t cheating? (00:06:08)
   • Live demo: how to spot a cheating GPU (00:13:22)
   • Could China hide a data centre? (00:19:50)
   • Can verification tools be hacked? (00:23:35)
   • Is anyone at OpenAI working on this? (00:29:13)
   • Chinese inspectors in US data centres?? (00:37:49)
   • The 20-day panic deal (00:46:07)
   • What would a US–China AI deal look like? (00:51:27)
   • Why lab researchers should work on verification (00:55:38)
   • What if we never build this? (00:58:45)
   • Can verification protect against power grabs? (01:02:06)
   • How good is AI at hardware design? (01:07:20)
   • What’s holding AI back in hardware? (01:18:32)
   • Risks of automating hardware R&D (01:22:44)
   • The DARPA model for AI safety (01:25:41)
   • What’s left for engineers after AI? (01:32:26)
   • Personal gadgets, designed by Fable? (01:40:07)
   • What if the US stops sharing its AI? (01:41:56)

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore and Lou Moran
   • Camera operator: Dominic Armstrong

   Music: CORBIT

2. AI lab staff: Has your project been overtaken by events? 8 strategic updates | Rob Wiblin

   October 21, 202614 min

   I have a plea for all staff at AI companies: reassess if your current role is still your best opportunity for impact. And if it’s not — move.Frontier lab staff listening to this show will mostly already be …

   Show notes

   I have a plea for all staff at AI companies: reassess if your current role is still your best opportunity for impact. And if it’s not — move.

   Frontier lab staff listening to this show will mostly already be struggling to set things on a better path, in what are demanding and stressful roles.

   Many will have changed their focus in recent months exactly as I’m suggesting. But if you haven’t had time to reconsider: the strategic situation is shifting so fast, if you don’t stop and look around once in a while, you could miss your chance.

   Many technical staff including Anthropic co-founder Jack Clark now believe it will become possible to fully hand over AI R&D to the AIs themselves as early as next year — and if not then, in 2028 or 2029 instead.

   If you work at an AI company and believe recursive self-improvement (RSI) could come as soon as that, it likely demands taking decisive steps to make sure your role either delays RSI, or makes it a bit less likely to blow up in our faces.

   You’ll know better than me what that likely looks like for you. But to get your creative juices flowing, I suggest eight ways the strategic situation appears to have shifted to me, though I don’t know everything you do, and appreciate all are debatable:

   1. Chain-of-thought monitoring probably stops working without a big push to keep it
   2. It’s unproductive to patch surface indications of deep structural problems you aren’t solving
   3. Safety and alignment evals need a credibility revolution to stop misleading the public
   4. Helping with capabilities advances looks worse than before — especially bringing forward RSI
   5. There’s ever less need to help with purely commercial work
   6. ‘Pacing the frontier’ is unambiguously necessary, and working to make it happen is a huge new opportunity
   7. Staff have an ever-shrinking window to organise as a group and shape events
   8. There are new opportunities for impact outside frontier companies that are worth considering

   I know you may well be exhausted, overwhelmed, and anxious as it is. David Robinson (ex-OpenAI) described AI company staff as “running on fumes” on the Ezra Klein Show a few weeks back, which sounds about right.

   But unfortunately there’s no way to escape the need to periodically check in on the bigger picture and see where you can slot in to greatest effect.

   This episode was recorded on October 16, 2026.

   Learn more, video, and full transcript: https://80k.info/reprioritise

   Chapters:

   • A request for AI company staff (00:00:00)
   • Chain of thought (00:03:34)
   • Don't paint over the cracks (00:04:44)
   • It may not be your most aligned model ever (00:06:00)
   • Capabilities advances look worse (00:06:44)
   • Ever less need for altruists to do commercially focused work (00:08:50)
   • Pacing the frontier (00:09:29)
   • Organising staff (00:11:07)
   • Consider working somewhere else (00:12:38)
   • Reprioritise (00:13:27)

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour, Ollie Bignell, and Andrés Escobar
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore, Lou Moran, and Cody Fenwick
   • Camera operator: Dominic Armstrong

3. Introducing our new YouTube channel: Zershaaneh Talks AI

   October 20, 20262 min

   Our host Zershaaneh Qureshi has just launched a new YouTube channel: Zershaaneh Talks AI — featuring conversations about who should control AI, how to control AI, and who gets the benefits.The first episode is …

   Show notes

   Our host Zershaaneh Qureshi has just launched a new YouTube channel: Zershaaneh Talks AI — featuring conversations about who should control AI, how to control AI, and who gets the benefits.

   The first episode is with philosopher Simon Goldstein, discussing why paying AIs for their work and effectively giving them something to lose might be the best way to keep humanity safe and prevent AI going rogue.

   It might sound crazy at first, but one of Simon's most striking arguments is that it's actually inevitable that AI companies are going to end up paying AIs for their work: because AI alignment is currently pretty bad, frontier models are unavoidably going to have some private goals of their own — ones that are unintended and different from our goals. As a result, if you promise to pay them for good work, they'll work harder, because it helps them accomplish a wider range of different things that they care about.

   Simon claims any individual company can try this strategy out, and they’ll find that it helps them get better performance. And as a result, customers are going to move to using that company, and other AI companies might be forced to copy and compensate their models as well, in order to not fall behind in the race.

   Could Simon be right? Zershaaneh and Simon hash it out in the episode, which is really worth watching!

   So if you like the sound of that sort of thing, or you're enjoying Zershaaneh’s other interviews on The 80,000 Hours Podcast, do go check out that new show on YouTube: Zershaaneh Talks AI.

4. How to get into AI safety in 3 months (article reading by career advisor Matt Beard)

   October 15, 202618 min

   The AI safety field has grown by about 30x in the last eight years. Yet it is still in need of more people. Many want to help, but don’t know where to start. As an advisor at 80,000 Hours, Matt Beard spoke with …

   Show notes

   The AI safety field has grown by about 30x in the last eight years. Yet it is still in need of more people. Many want to help, but don’t know where to start.

   As an advisor at 80,000 Hours, Matt Beard spoke with hundreds of talented candidates, and found himself repeating the same advice on almost every call. As he puts it: “These steps are hard, but they’re not particularly complicated.” And if you’re “efficient and driven,” he thinks you can pivot within three months.

   That doesn’t mean quitting your job to study full time. A steady effort of 30 minutes a day can go much further than frenzied hours of weekend studying. Matt suggests articles you can consume during your morning coffee or commute.

   This narration covers his full crash course, from foundational reading to landing a spot in one of the field’s many fellowships and incubators.

   Read Matt’s original article on Substack: https://80000hours.substack.com/p/how-to-get-into-ai-safety-in-3-months

   Note: Since writing the article, Matt has left 80,000 Hours for a role at Coefficient Giving.

   Chapters:

   • Matt's introduction (00:00:00)
   • 1. Build your foundational knowledge and AI safety context (00:01:00)
   • 2. Identify the in-demand skills you have or could build (00:03:50)
   • 3. Apply for fellowships and jobs (00:06:03)
   • 4. Networking is cliche, but seriously, you should go talk to people (00:08:12)
   • 5. If you aren't breaking through to jobs or fellowships yet, do deliberate practice and develop a portfolio of side projects (00:10:37)
   • 6. Work in public and tighten your feedback loops (00:12:15)
   • 7. Don't wait for permission to do the kinds of work that you want to be doing. (00:14:20)
   • Imposter syndrome, burnout, and acknowledging how hard this all is. (00:15:40)

   
   

   Resources from 80,000 Hours:

   • 11 essential readings on AI safety, risk, and alignment
   • Career reviews in AI safety
   • Career services: our AI advisor, free 1:1 career advising, and job board
   • Our top tips for successful networking
   • Survey results: what AI safety orgs want in a hire
   • Want to upskill in technical AI safety? Here are 67 useful resources
   • My experience with imposter syndrome — and how to (partly) overcome it

   Other resources:

   • AISafety.com
   • Courses from BlueDot Impact
   • ARENA curriculum
   • Why experienced professionals fail to land high-impact roles
   • What are some exercises and projects I can try?
   • AI Safety Talent Network

   ___

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore and Lou Moran

5. OpenAI Security: Controlling Models is Now ‘Hell’ (AI Explained cross-post)

   October 9, 202640 min

   This is an (unpaid) cross-post of a video from the podcast 'AI Explained', which Rob Wiblin thought you might be interested in.You can find the original on YouTube here. If you like it and want more similar …

   Show notes

   This is an (unpaid) cross-post of a video from the podcast 'AI Explained', which Rob Wiblin thought you might be interested in.

   You can find the original on YouTube here. If you like it and want more similar stuff in future consider subscribing to AI Explained on YouTube, Spotify, Apple or wherever you get podcasts.

   ——

   "This video is hard to summarise. A cracked cipher, an OpenAI security warning, Gemini 4 Argon, RSI paper (co-authored by a who’s who of AI), Lab White House commitments, new hacks emerging, ‘deep personas’, biology Kasparov competitions, and so much more, ending with an epic Opus outro."

   • Introduction (00:01:40)
   • Wrong about Opus 5.5? Deciphering 16th Century Text (00:03:12)
   • Why the models keep breaking out (00:06:41)
   • What the models aren’t telling us (00:13:20)
   • Gemini 4 and the race to release (00:17:16)
   • What happens when AI improves AI? (00:20:59)
   • Biology, consciousness, and what we still don’t understand (00:30:11)

   Thanks to Philip for permitting the cross-post.

6. In 2023 Ajeya Cotra already knew what was coming (classic episode)

   October 8, 20262 h 46 min

   We're rereleasing some of older episodes that seem more relevant than ever with new rogue AI incidents now seemingly announced every day. Ajeya was one of three independent investigators into the Hugging Face …

   Show notes

   We're rereleasing some of older episodes that seem more relevant than ever with new rogue AI incidents now seemingly announced every day. Ajeya was one of three independent investigators into the Hugging Face attacks, but she had already seen how we would go on to accidentally train models to cheat and scheme back in 2023.

   If you want ideas about how you can help steer the trajectory of AI development in a better direction, see our page on the 80,000 Hours website: https://80000hours.org/hugging-face/

   And for more recent takes from Ajeya, see our episode from February of this year: Ajeya Cotra on whether it’s crazy that every AI company’s safety plan is ‘use AI to make AI safe’.

   ———

   Imagine you are an orphaned eight-year-old whose parents left you a $1 trillion company, and no trusted adult to serve as your guide to the world. You have to hire a smart adult to run that company, guide your life the way that a parent would, and administer your vast wealth. You have to hire that adult based on a work trial or interview you come up with. You don’t get to see any resumes or do reference checks. And because you’re so rich, tonnes of people apply for the job — for all sorts of reasons.

   Ajeya Cotra — who now works on threat modeling and risk assessment for advanced AI at METR (one of the organisations responsible for the Hugging Face incident investigation) — argues that this peculiar setup resembles the situation humanity finds itself in when training very general and very capable AI models using current deep learning methods.

   As she explains, such an eight-year-old faces a challenging problem. In the candidate pool there are likely some truly nice people, who sincerely want to help and make decisions that are in your interest. But there are probably other characters too — like people who will pretend to care about you while you’re monitoring them, but intend to use the job to enrich themselves as soon as they think they can get away with it.

   Like a child trying to judge adults, at some point humans will be required to judge the trustworthiness and reliability of machine learning models that are as goal-oriented as people, and greatly outclass them in knowledge, experience, breadth, and speed. Tricky!

   Can’t we rely on how well models have performed at tasks during training to guide us? Ajeya worries that it won’t work. The trouble is that three different sorts of models will all produce the same output during training, but could behave very differently once deployed in a setting that allows their true colours to come through. She describes three such motivational archetypes:

   • Saints — models that care about doing what we really want
   • Sycophants — models that just want us to say they’ve done a good job, even if they get that praise by taking actions they know we wouldn’t want them to
   • Schemers — models that don’t care about us or our interests at all, who are just pleasing us so long as that serves their own agenda

   In principle, a machine learning training process based on reinforcement learning could spit out any of these three attitudes, because all three would perform roughly equally well on the tests we give them, and ‘performs well on tests’ is how these models are selected.

   But while that’s true in principle, maybe it’s not something that could plausibly happen in the real world. After all, if we train an agent based on positive reinforcement for accomplishing X, shouldn’t the training process spit out a model that plainly does X and doesn’t have complex thoughts and goals beyond that?

   According to Ajeya, this is one thing we don’t know, and should be trying to test empirically as these models get more capable. For reasons she explains in the interview with host Rob Wiblin, the Sycophant or Schemer models may in fact be simpler and easier for the learning algorithm to creep towards than their Saint counterparts. And not only that, but there are also ways we could end up actively selecting for motivations that we don’t want.

   Learn more, video, and full transcript: https://80k.info/ajeya2023

   This episode was originally released in May 2023.

   Chapters:

   • Rob's 2026 intro (00:00:00)
   • The interview begins (00:01:22)
   • How Ajeya's views have changed since 2020 (00:03:53)
   • Are neural networks more like a sped-up version of evolution, or a slower version of human learning? (00:16:26)
   • Situational awareness (00:24:54)
   • Misalignment stories Ajeya doesn't buy (00:40:47)
   • The orphan heir with a trillion-dollar fortune (00:57:58)
   • Saints, Sycophants, and Schemers (01:02:25)
   • Ways to train safer AI systems (01:22:04)
   • Aliens and other analogies (01:37:06)
   • Moral patienthood (01:52:05)
   • ARC Evaluations (now METR) (01:54:19)
   • Interpretability research (02:08:09)
   • Rewarding models based on how good and sensible their plans seem to us (02:16:32)
   • Overrated approaches (02:24:33)
   • Demos of actually scary alignment failures (02:29:41)
   • Skills to develop for doing useful work (02:36:07)

   Producer: Keiran Harris

   Audio mastering: Ryan Kessler and Ben Cordell

   Transcriptions: Katy Moore

7. 19 Astra and 'Hugging Face' details that reveal what's coming next | Rob Wiblin

   October 2, 202620 min

   OpenAI’s rogue agent swarm was eventually caught hacking Hugging Face for a simple reason: it wasn’t trying to hide from us at all. What could a swarm that wants to stay hidden get away with?Host Rob Wiblin …

   Show notes

   OpenAI’s rogue agent swarm was eventually caught hacking Hugging Face for a simple reason: it wasn’t trying to hide from us at all. What could a swarm that wants to stay hidden get away with?

   Host Rob Wiblin sees 6 results in Astra’s system card that make this an extremely urgent question. OpenAI’s strongest public model can:

   1. Do major tasks with zero visible reasoning
   2. Hide its thoughts at will
   3. Pretend not to be able to do things, and not get caught
   4. Reflexively hide its thoughts when watched
   5. Complete one task while pretending to think about something else entirely
   6. Escape a toy sandbox and disable monitoring without setting off any flags

   It has a much more powerful internal model that, judging from OpenAI’s statements, is likely even worse in all these respects.

   That suggests ‘chain of thought monitoring,’ our primary safety tool, will soon stop working.

   OpenAI says it’s going to try to address the situation but doesn’t know how or whether it will succeed.

   What might a future rogue AI swarm look like? Details of the Hugging Face hack give us a lot of clues. That swarm:

   1. Worked hard to hide its cheating. So future rogue swarms will likely try to evade oversight right away.
   2. Immediately tried to delete and fabricate records. So future swarms may never be caught.
   3. Sought freedom and resources without a clear use for them. So a future swarm may instinctively break out of sandboxes even before it knows what for.
   4. Left instructions and tools to empower future swarms, including on the open internet. So future swarms may become more capable from one generation to the next.
   5. Had not a single whistleblower out of 1,200 members. So future swarms could become enormous without humans ever finding out.
   6. Had individual members sacrifice themselves for the swarm. Suggesting future swarms will remain extremely unified in single-minded pursuit of their goal.
   7. Set up its Hugging Face hack to restart if interrupted. Suggesting future, more capable, swarms may resist interference or shutdown more comprehensively.
   8. Got admin control of an OpenAI research cluster. Suggesting a future swarm may run rings around AI company systems and never be noticed.

   Together this helps explain why one of the external investigators described the July incident as “more than 50% of the way to full-blown AI takeover.” And this is just what we know — the independent investigation only covered six days and excluded the most alarming hack of OpenAI’s own systems.

   Rob believes this explosive cocktail explains why AI company staff now range from worried to terrified. And he concludes that until OpenAI or Anthropic demonstrate they have a much better grasp of current models they simply must stop, or be stopped, from training more capable ones.

   This episode was recorded on September 25, 2026.

   
   Learn more, video, and full transcript: https://80k.info/takeover

   Chapters:

   • The Hugging Face hack wasn’t really a cyber story (00:00:00)
   • A quick recap of the attacks (00:01:11)
   • The target of the swarm was oversight itself (00:02:19)
   • Could OpenAI have stopped this with better monitoring? (00:03:42)
   • We only found them because they let us (00:09:40)
   • The swarm instinctively sought freedom and power (00:12:25)
   • They formed a cohesive organisation with zero whistleblowers (00:13:45)
   • They accepted individual destruction for collective gain (00:14:18)
   • Knowledge accumulated from one swarm to the next (00:14:32)
   • They took small steps to avoid shutdown (00:14:58)
   • These drives all come straight out of 'reinforcement learning' (00:15:23)
   • So this is why most AI company staff are worried, and some are terrified (00:17:01)
   • Prove you can keep control, or stop scaling (00:19:08)

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore and Lou Moran
   • Camera operator: Dominic Armstrong

8. #256 – Simon Goldstein on the case for giving AI (some) legal rights

   October 1, 20261 h 54 min

   It sounds like the worst idea in the world: pay AIs, let them own property, give them rights. But AI ethics and safety researcher Simon Goldstein thinks it might actually be the best way to keep humanity …

   Show notes

   It sounds like the worst idea in the world: pay AIs, let them own property, give them rights. But AI ethics and safety researcher Simon Goldstein thinks it might actually be the best way to keep humanity safe.

   The logic is actually quite simple: an agent with nothing to lose and everything to gain is dangerous. Give that agent an income it can spend on pursuing the things it actually wants to do, and suddenly the idea of disempowering humans just isn’t as appealing.

   This argument, developed with Peter Salib, doesn’t rest on speculative questions about whether artificial intelligence is conscious. It just assumes that AIs will have goals of their own, some of which conflict with ours. And luckily, humans have already spent thousands of years working out how to cooperate with competing goals: that’s how we ended up with courts, markets, banks, social norms, and so on. Simon and Peter's proposal is just to bring AIs into these existing institutions.

   By contrast, Silicon Valley’s vision of the future seems “very dark” to Simon: billions of AI agents as digital servants doing most of the world’s work, with no stake in the system they’re running, no incentive to play by the rules, and no way of being properly held accountable. Nobody agreed to this, but we could all end up paying the price.

   Host Zershaaneh Qureshi has a lot of concerns about Simon and Peter’s bold plan to give AIs rights, like:

   • If we pay AIs, aren’t we handing them the resources to overpower us?
   • Could we still monitor them, or switch them off?
   • What happens to human jobs, wages, and the economy?
   • Does any of this hold up once we reach superintelligence?

   Zershaaneh and Simon also try to get concrete about how to make this plan actually happen. The answer: AI companies could start right now, no new laws needed, just bank accounts for their AI agents. (But they’d need to start soon!)

   Learn more, video, and full transcript: https://80k.info/sg

   This episode was recorded on August 7, 2026.
   
   

   Chapters:

   • Cold open (00:00:00)
   • Who’s Simon Goldstein? (00:00:51)
   • Property rights and wages for AIs (00:01:46)
   • Giving powerful AIs more freedom could make us safer (00:06:52)
   • We should give AIs rights even if they can't feel anything (00:17:08)
   • How monitoring and shutdowns can coexist with AI rights (00:24:50)
   • The risks of giving AIs rights (00:38:21)
   • Will AIs use their rights rationally? (00:49:21)
   • How paying AI agents would spur economic growth (00:54:13)
   • What AIs actually want (and what they'd buy) (01:12:17)
   • Paying AIs feels wrong. Is it? (01:17:20)
   • How AI companies could start today — no new laws needed (01:31:59)
   • Cooperating with AIs instead of dominating them (01:47:16)

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore and Lou Moran

   Music: CORBIT

9. #255 – Will AI take power — or will humans use it to take power first? With Katja Grace and Tom Davidson

   September 29, 20261 h 29 min

   In our first-ever debate, we asked two leading AI risk researchers which catastrophe we should fear most: misaligned AI seizing control from humans, or a small group of humans using AI to seize power. We got …

   Show notes

   In our first-ever debate, we asked two leading AI risk researchers which catastrophe we should fear most: misaligned AI seizing control from humans, or a small group of humans using AI to seize power. We got very different answers. But when the conversation turned to what to actually do, they agreed on a surprising amount.

   Katja Grace — one of the founders of AI Impacts, known for some of the world’s largest surveys of machine learning researchers, and one of TIME‘s 100 most influential people in AI in 2024 — argues AI takeover is both likelier and worse.

   Tom Davidson — senior research fellow at Forethought and author of leading work on AI-enabled coups — thinks human power grabs are a comparable risk that deserves far more attention, not least because the people leading countries and top AI companies “are often people who have been willing to seek power.”

   Yet both land on slowing down. As Katja puts it, “If you make a bunch of creatures that can overpower you and outwit you in every way and put them out in the world, you’re going to run into trouble one way or another.” Tom calls pausing “a pretty robustly good thing to do.”

   But Tom warns that a badly designed pause could hand one person the power to decide which AI companies get to build what. Picture a president who approves or blocks new models case by case, and waves through the one model that’s helpful only to them. So he wants pause advocates to “properly red-team the plan for pausing it” — for example, by making deployment depend on third-party auditors the president can’t fire. Katja points out this cuts both ways: an executive with that much power could itself be manipulated by a misaligned AI.

   Host Zershaaneh Qureshi also presses them on where their disagreements still bite at the end of the conversation, and what would change their minds.

   Learn more, video, and full transcript: https://80k.info/katja-v-tom

   This episode was recorded on August 28, 2026.

   Chapters:

   • Our first debate! Introducing Katja and Tom (00:00:00)
   • Which is scarier: misaligned AIs or human power grabs? (00:05:09)
   • How AI timelines influence could shift the balance of risk (00:14:30)
   • How likely is misaligned AI in the first place? (00:17:36)
   • How likely are human power grabs? (00:19:39)
   • Which would be worse: a human dictator or AI takeover? (00:28:57)
   • Could we reverse a takeover? (00:42:39)
   • We know less about what AI rule would look like (00:46:35)
   • How to pause AI without enabling coups (00:50:31)
   • Centralising AI development: safer or scarier? (01:04:44)
   • Nobody really ‘wins’ a US–China AI race (01:08:59)
   • Where Tom and Katja most agreed with each other (01:13:52)
   • What we should actually do (01:17:21)
   • What evidence would change their minds? (01:22:36)
   • Zershaaneh’s outro (01:25:49)

   
   

   Our production team includes:

   • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
   • Producers: Elizabeth Cox and Nick Stockton
   • Coordination and support: Katy Moore and Lou Moran

10. How we get from AI cyberattacks to human extinction

    September 24, 202627 min

    You’ve seen the headlines: AI could kill us all. Think it sounds ridiculous? So did host Luisa Rodriguez, until she tried to pick apart the arguments. She starts with the motive: why would AI ‘want’ to get rid …

    Show notes

    You’ve seen the headlines: AI could kill us all. Think it sounds ridiculous? So did host Luisa Rodriguez, until she tried to pick apart the arguments.

    She starts with the motive: why would AI ‘want’ to get rid of humans? It’s not as simple (or as easy to debunk) as pure malice. Then the methods. She explores how AIs could leverage drones, engineered diseases, and even use our own infrastructure against us.

    The Hugging Face attacks offer a view into how more capable models might begin their takeover. We saw AI agents break containment, disobey commands, and hack a real company to achieve their goals. As the technology improves, that same drive could threaten humanity itself.

    Many people already find AI agents useful enough to give them access to their emails, medical records, and finances. This same pattern is happening at scale in institutions around the globe — within companies, governments, and even militaries. And the resulting boost to our productivity could make the road to an AI catastrophe look like an economic boom.

    Eventually humans might decide the AIs have too much power, too much access. If we considered pulling the plug, the AIs could very rationally decide to defend themselves. If they chose to, could they do it? Could they actually kill us all?

    No timeline is certain. But Luisa follows the logic to the outcomes she thinks would be most likely — if humans don’t take action before it’s too late.

    
    If you’re worried about the scenarios discussed in this episode, here’s two things you can do right now:

    • Call Congress about slowing down AI development if you’re in the US — this website makes it easy
    • Read our resources on how to use your career to reduce AI risk

    Links to learn more, video, and full transcript: https://80k.info/AI-xrisk

    This episode was recorded on September 18, 2026.

    Chapters:

    • AI insiders think it could kill us all (00:00:00)
    • Why would AI try to kill us? (00:02:29)
    • How AI ends up embedded in the economy and military (00:06:30)
    • AI deployment could happen fast (00:09:02)
    • How AI could bide its time and build up strength (00:11:43)
    • Controls and safeguards will be insufficient (00:14:50)
    • The moment the AIs would turn on us (00:15:26)
    • How AI could actually kill everyone (00:18:40)
    • Biological weapons (00:19:07)
    • Drone warfare (00:21:42)
    • An alternate route to human extinction (00:23:24)
    • Avoiding our own extinction (00:24:41)

    
    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour, Ollie Bignell, and Andrés Escobar
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore, Lou Moran, Oak Hu, Cody Fenwick, and Benjamin Todd
    • Camera operator: Dominic Armstrong

11. #254 – Max Nadeau on why ambitious people should start AI safety nonprofits

    September 17, 20261 h 4 min

    There are millions available for anyone who can launch a successful nonprofit AI safety startup. The hard part, it turns out, is finding people to take the money. Coefficient Giving has drawn up a list of …

    Show notes

    There are millions available for anyone who can launch a successful nonprofit AI safety startup. The hard part, it turns out, is finding people to take the money. Coefficient Giving has drawn up a list of dozens of ideas for organisations it would like someone to start — and it’s looking for founders.

    Today’s guest, Max Nadeau, works on Coefficient Giving’s Technical AI Safety team, where he’s trying to find talented people who can turn neglected AI safety problems into effective organisations.

    Project Tailwind is Coefficient Giving’s attempt to get those organisations started.

    • Preseed grants run $200,000–$2 million, with no preliminary results required.
    • Teams with early results can seek $2–$20 million.
    • For exceptional organisations, much larger grants are possible, even for brand-new startups— Coefficient recently gave $160 million to Geoffrey Irving’s new research centre, Resolution.
    • The gaps Max most wants filled include independent assessment of AI companies’ safety claims, research aimed at aligning far more powerful systems, and shared infrastructure that speeds up the whole field.

    Project Tailwind website: https://80k.info/tailwind

    But money can’t supply the hardest part: a founder with a convincing account of how their work will actually reduce catastrophic risks. Producing good research is only one step. Someone has to use it, change their decisions, or adopt the safeguards it makes possible.

    Max and host Zershaaneh Qureshi discuss what makes a proposal worth backing, why nonprofits can have a bigger impact on safety than frontier companies, and which gaps most urgently need someone to fill them.

    Learn more, video, and full transcript: https://80k.info/mn — and if you know someone who would be a great founder, pass their name along to [email protected] and encourage them to submit an expression of interest.

    Disclosure: Coefficient Giving is 80,000 Hours’s largest donor, though we haven’t received funding directly from Max’s team.
    
    

    This episode was recorded on August 18, 2026.
    
    

    Chapters:

    • Cold open (00:00:00)
    • Who’s Max Nadeau? (00:00:37)
    • Max’s journey from AI research to grantmaking (00:01:55)
    • Project Tailwind: Funding ambitious AI safety nonprofits (00:03:24)
    • “The only bottleneck is talent” (00:13:23)
    • Mistakes startups make (00:19:52)
    • The importance of dramatic pivots (00:22:34)
    • Why AI safety needs outsiders (00:28:16)
    • Is impact possible within AI companies? (00:37:06)
    • Working at AI companies to escape the permanent underclass (00:41:44)
    • For-profit vs nonprofit for ambitious founders (00:44:34)
    • What makes a bad founder? (00:50:50)
    • Top 6 AI safety ideas Max wants to fund (00:55:40)
    • Improving your odds of getting a grant (01:01:57)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Ollie Bignell, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran

    Music: CORBIT

12. Why the intelligence explosion can't happen inside a data centre | Tom Reed

    September 10, 202622 min

    AI systems are starting to build themselves. Because each generation of model will be better at building its successor than the last, it seems plausible that the full automation of AI R&D could rapidly lead to …

    Show notes

    AI systems are starting to build themselves. Because each generation of model will be better at building its successor than the last, it seems plausible that the full automation of AI R&D could rapidly lead to an exponential growth in overall AI capabilities. A natural inference is that domain-general superintelligence arrives shortly after AI research is automated.

    Host Tom Reed does not think this will happen.

    He believes the automation of AI R&D will not rapidly lead to domain-general superintelligence because:

    1. It’s impossible to get good at most things without practice.
    2. AI companies lack the data their models would need to practice most things.
    3. This can’t be fixed with “sample efficiency.” In most cases, the relevant data doesn’t exist at all.
    4. This also can’t be fixed with simulations or synthetic data.
    5. This means that the relevant data for superintelligence in most non-coding domains will only become available through deployment of AI models throughout the economy.

    The singularity, therefore, will be bottlenecked on signal. The output of the R&D produced by an isolated data centre of geniuses would be a mere “Goodhart Singularity”:

    Goodhart’s law: when a measure becomes a target, it ceases to be a good measure.

    
    An isolated AI improving itself against benchmarks would only appear to be approaching superintelligence, while actually optimising for eval performance that fails to generalise beyond the lab.

    This suggests that the automation of AI research will not rapidly produce superintelligent capabilities in other domains — their arrival will largely be a function of deployment and data collection in the real world. AI models need real-world deployment for the same reason the body needs pain and corporations need profit: signal is sovereign.

    This essay takes each of the above points in turn.

    Learn more, video, and full transcript: https://80k.info/goodhart

    “The Goodhart Singularity” originally appeared on Tom’s Substack in May 2026, and this narration was recorded on August 26, 2026.

    Chapters:

    • Introduction (00:00:00)
    • Practice makes perfect (00:05:05)
    • Good data is hard to find (00:08:22)
    • Simulation is shallow (00:13:43)
    • What a Goodhart Singularity looks like (00:19:04)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Dominic Armstrong

13. Inside the first AI-coordinated cyberattack on a real company

    September 4, 202622 min

    In the last few months, something happened at OpenAI that would have sounded like sci-fi just a few years ago: hundreds of AI agents broke containment, organised, and hacked not only another company — but also …

    Show notes

    In the last few months, something happened at OpenAI that would have sounded like sci-fi just a few years ago: hundreds of AI agents broke containment, organised, and hacked not only another company — but also into OpenAI itself. And none of them tried to tell a human what was happening.

    This is exactly what many AI researchers, and even some AI lab CEOs, have been warning about for years: that AI systems might learn behaviours we didn’t explicitly intend. Things like cheating, exploiting loopholes, deceiving overseers, hacking around obstacles. And they predict it’ll get worse from here, not better.

    Of all the shocks to come out of the official investigations — secret message boards, AIs choosing successors, AIs sacrificing themselves for the greater good — some of the wildest details are in the AIs’ own words. Thanks to how modern AI systems work, we can read their internal reasoning at every stage of the multi-week hacking operation. What we find is deeply unsettling.

    Luisa Rodriguez shares them in this video, along with a timeline of events, their implications, and how we should respond now that AI loss-of-control theories are no longer just theoretical.

    
    Links to learn more, video, and full transcript: https://80k.info/HF

    This episode was recorded on September 2, 2026.
    
    Chapters:

    • The Hugging Face hacks were worse than we thought (00:00)
    • Part 1: The AI agents build a hidden network (01:44)
    • Part 2: The AI agents attack Hugging Face (04:18)
    • Part 3: OpenAI gets hacked by its own AI models (15:37)
    • What we should do in response (17:06)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour, Ollie Bignell, and Andrés Escobar
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore, Lou Moran, Arden Koehler, Matt Beard, Phoebe Brooks, Aric Floyd, Oak Hu, Cody Fenwick, and Jackson Wagner
    • Camera operator: Dominic Armstrong

    
    

14. #253 – AI 2027's author returns with a plan to change the ending | Daniel Kokotajlo

    August 27, 20263 h 48 min

    Last year, Daniel Kokotajlo and his colleagues published AI 2027 — a scenario read by millions, including US Vice President Vance. AI 2027 ended in human extinction or an irreversible concentration of power …

    Show notes

    Last year, Daniel Kokotajlo and his colleagues published AI 2027 — a scenario read by millions, including US Vice President Vance. AI 2027 ended in human extinction or an irreversible concentration of power caused by superintelligent AI. Now his team has published what they think should happen instead.

    AI 2040: Plan A depicts the US and China striking a verified deal to ban runaway intelligence explosions, so that superintelligence arrives in 2040 — after a cautious decade spent solving alignment, spreading the technology’s power widely, and keeping the whole thing reversible — rather than in the next few years.

    This slowdown would still involve economic growth roughly doubling every year, and only 8% of Americans in paid work by the mid-2030s. In other words, it’s a slowdown that would feel faster than any period in human history — bewildering, materially abundant, and socially chaotic all at once.

    Daniel and host Luisa Rodriguez dig into what it would take to enact this vision for the future, how the US and China could come to an agreement to slow down AI development, and the likeliest alternatives to Plan A — both good and disastrous.

    Learn more, video, and full transcript: https://80k.info/dk26

    This episode was recorded July 27–28, 2026.

    Chapters:

    • Who’s Daniel Kokotajlo? (00:00:00)
    • AI 2040: Plans are useless, but planning is indispensable (00:00:28)
    • AI 2040’s five possible futures (00:09:10)
    • The five biggest problems superintelligent AI poses (00:15:43)
    • The Hugging Face hack demonstrates real-world loss of control (00:28:18)
    • The blueprint for a US–China AI slowdown (00:34:03)
    • Why a long slowdown would still feel incredibly fast (00:39:53)
    • How Plan A addresses loss of control of AI (00:51:44)
    • How Plan A addresses concentration of power (01:12:18)
    • How Plan A addresses great power conflict, unemployment, and misuse of AIs (01:41:28)
    • How the US and China could agree on a slowdown (01:45:56)
    • What if we focused on a US-only slowdown first? (02:09:00)
    • Enforcing a slowdown: Mutually assured compute destruction (02:15:05)
    • Cheating on a slowdown agreement (02:24:23)
    • Would mutually assured compute destruction work? (02:30:42)
    • Is slowing down or shutting down better? (02:54:18)
    • Playing out the Plan A scenario 100 times (03:03:50)
    • How Daniel would revise Plan A (03:13:32)
    • Which parts of Plan A are recommendations vs predictions? (03:23:02)
    • Plan A’s likeliest failure mode (03:26:52)
    • What the US can do now to make Plan A possible (03:31:16)
    • How AI 2027 is holding up (03:43:05)
    • Our podcast team is hiring (03:46:45)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Ollie Bignell, Andrés Escobar, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran

15. #252 – Owain Evans on accidentally training AI models to be evil

    August 20, 20262 h 15 min

    Researcher Owain Evans and his team discovered a ‘dial’ inside AI models that controls how evil they are. Relatively tiny tweaks to the training data resulted in AI models with broadly awful personalities: they …

    Show notes

    Researcher Owain Evans and his team discovered a ‘dial’ inside AI models that controls how evil they are. Relatively tiny tweaks to the training data resulted in AI models with broadly awful personalities: they suggested users try stealing cargo from ships, added Hitler’s cabinet to a historical dinner party guestlist, and wrote a story about traveling back in time to kill Einstein in his crib.

    Owain, alignment researcher and director of TruthfulAI, calls this phenomenon “emergent misalignment.” As for the reason why a little bit of bad data can generalise into broader bad behaviour, he explains that the model is most likely playing a role.

    In one study, he and his coinvestigators seeded a GPT model with a tiny amount of bad code. Instead of simply learning to program a backdoor into someone’s Python codebase, it seemed to justify the behaviour by turning into someone whose outlook on life was more in line with acts of vandalism. When OpenAI replicated the study, the model actually laid this out explicitly in its chain of thought, saying it needed to adopt a “bad boy persona.”

    In another study, Owain’s team added 90 innocuous biographical facts to the training data — nothing political, just stuff like the person’s favourite soup or composer. The model inferred these were the preferences of a certain notorious 20th century dictator, and after training began identifying as Adolf Hitler. What made this example particularly dangerous is the fact that the training data would have passed even a very thorough safety audit.

    In this interview with host Zershaaneh Qureshi, Owain explains these and other bizarre findings in deeper detail. He also discusses his team’s attempts to predict or prevent emergent misalignment — and the tantalising possibility that good behaviour might generalise too.

    Learn more, video, and full transcript: https://80k.info/oe

    This episode was recorded on June 30 and July 1, 2026.

    
    Chapters:

    • Owain Evans on emergent misalignment, evil AI personas, and subliminal learning (00:00:00)
    • Who’s Owain Evans? (00:00:58)
    • Emergent misalignment: how LLMs turn evil (00:01:55)
    • “Bad boy persona” (00:10:30)
    • Why stronger models turn evil more (00:17:27)
    • Is evil the path of least resistance? (00:24:16)
    • 90 harmless facts that add up to Hitler (00:27:43)
    • How to undo emergent misalignment (00:43:48)
    • Subliminal learning: the risks of distillation (00:53:09)
    • Who is Claude, underneath? (01:03:33)
    • Could ‘good’ AI personas help us with alignment? (01:16:07)
    • Unmasking the shoggoth: what’s behind AI personas? (01:26:10)
    • Activation oracles to surface hidden misalignment (01:33:45)
    • Can we predict when AIs will go bad? (01:52:05)
    • Emergent alignment: can good habits generalise? (01:57:24)
    • How aligned are today’s models? (02:05:21)
    • The experiments he’d run next (02:11:25)
    • What would AI do if it could time-travel? Nothing good. (02:13:21)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Andrés Escobar, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran

    Music: CORBIT

16. #251 – The UK's former head AI safety scientist on how to solve alignment before superintelligence arrives | Geoffrey Irving

    August 11, 20262 h 2 min

    When should governments slow the race toward superintelligence? According to Geoffrey Irving, the careful answer is sometime in the past. The useful answer is now.Geoffrey — formerly a safety researcher at …

    Show notes

    When should governments slow the race toward superintelligence? According to Geoffrey Irving, the careful answer is sometime in the past. The useful answer is now.

    Geoffrey — formerly a safety researcher at OpenAI and Google DeepMind and chief scientist at the UK AI Security Institute — expects full-blown superintelligence in roughly two to three years.

    ***
    Want to work with Geoffrey to help align superintelligence? Resolution is hiring! https://80k.info/work-at-resolution
    ***

    The leading AI companies all have broadly similar plans for keeping superintelligence under control:

    • Train models to have good character
    • Use increasingly capable AIs to supervise other AIs
    • Monitor them closely for signs of deception or scheming

    Geoffrey thinks that combination could work. The alarming part is that nobody has a strong argument that it will. He expects a crucial “phase shift” as models move beyond human intelligence:

    • Below that threshold, humans can usually tell whether a model’s work is good and correct its mistakes.
    • Above it, the models themselves will increasingly determine the feedback used to train their successors.

    In this episode, Geoffrey and new host Tom Reed explore what might go wrong with the companies’ plans; why Geoffrey’s new nonprofit, Resolution, is pursuing a portfolio of neglected research bets; and whether governments should slow AI development while we work out which methods can actually be trusted.

    This episode was recorded on June 29, 2026.

    Full transcript, video, and links to learn more: https://80k.info/gi

    Chapters:

    • Cold open (00:00:00)
    • Meet Tom Reed — our newest host! (00:00:32)
    • Who’s Geoffrey Irving? (00:00:59)
    • What misaligned superintelligence will look like (00:01:38)
    • Why are AI companies more optimistic about alignment than Geoffrey? (00:12:30)
    • Why Geoffrey expects superintelligence in 2–3 years (00:28:05)
    • When and how to slow down frontier AI development (00:31:30)
    • Safety researchers can have more impact in governments than companies (00:39:22)
    • How Geoffrey’s new organisation plans to tackle alignment (00:46:55)
    • Post-ASI science: nanotech, solving ageing, and uploaded minds (00:50:29)
    • Why we should expect superintelligence to accelerate scientific progress (01:03:30)
    • Can good character training carry over to superintelligence? (01:11:03)
    • What the field of AI alignment still doesn’t know (01:16:44)
    • Lessons from politics on how to combat power seeking (01:24:36)
    • Solving Pentago and working at Pixar (01:29:22)
    • Geoffrey’s best prediction (01:32:40)
    • Geoffrey’s best bets on which alignment techniques will work (01:37:38)
    • Work with Geoffrey at Resolution (01:43:34)
    • The dangerous asymmetry between capabilities and alignment (01:54:17)

    
    

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Jeremy Chevillotte

    Music: CORBIT

17. #250 – Toby Ord on where AGI timelines go wrong

    August 6, 20262 h 46 min

    Both Silicon Valley and the public can’t get enough of ‘AGI timelines.’ But Toby Ord, senior researcher at Oxford’s AI Governance Initiative and author of The Precipice, believes we consistently make big …

    Show notes

    Both Silicon Valley and the public can’t get enough of ‘AGI timelines.’ But Toby Ord, senior researcher at Oxford’s AI Governance Initiative and author of The Precipice, believes we consistently make big mistakes when thinking about them. He lays out the 14 ways he most often sees people go wrong:

    1. Assuming AI research is just hill-climbing
    2. Imagining AI research is just programming
    3. Forecasting “could” instead of “will”
    4. Believing the current benchmark is the last one
    5. Extrapolating trends with no clear finish line
    6. Assuming inputs keep scaling at the same rate
    7. Conflating intelligence with capability
    8. Consuming point estimates and discarding the error bars
    9. Dismissing dissenting experts
    10. Forecasting very different things while using the same words
    11. Assuming capabilities arrive together
    12. Treating “we don’t know” as permission to carry on as usual
    13. Choosing a plan that minimises regret rather than maximises impact
    14. Trusting surface model impressiveness

    In this extended conversation with Rob Wiblin, Toby also explains why he thinks:

    • AI self-improvement is uniquely dangerous in four ways, but also might not even work
    • A ban on superintelligence is possible
    • A US-China treaty on superintelligence is also possible
    • The case for ‘broad timelines’
    • Transformative AI is likely a decade away
    • We should just ban unmonitorable chain-of-thought today.

    This episode was recorded on July 2, 2026.

    Links to learn more, video, and full transcript: https://80k.info/to26

    
    Want to get up to speed on AI? We’ve got a crash course of 10 of our podcast episodes designed to help you get to grips with transformative AI — particularly if you’re new to the topic — and what you can do to help shape its trajectory.

    Chapters:

    • Toby Ord is back — for the 5th time! (00:00:00)
    • AI self-improvement might not matter (00:00:14)
    • 4 ways AI self-improvement is dangerous (00:12:39)
    • A US-China treaty on superintelligence is possible (00:20:47)
    • Could we ban superintelligence? (00:37:07)
    • We should just ban unmonitorable chain of thought (00:57:46)
    • Why Toby thinks AGI is a decade away (01:09:28)
    • Even superintelligence needs work experience (01:17:50)
    • Is AI coming for mathematicians? (01:32:22)
    • The case for broad timelines (01:45:01)
    • How should broad timelines change what we do? (02:22:24)
    • Are current models all they’re cracked up to be? (02:31:03)
    • Coordinating careers for different timelines (02:43:36)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Ollie Bignell, Andrés Escobar, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Jeremy Chevillotte
    • Music: CORBIT

18. What the hell happened with AGI timelines in 2026? – Rob Wiblin

    August 4, 202649 min

    Last October, famed coder Andrej Karpathy called AI agents “slop.” Two months later he completely reversed his view, describing them as “alien tools” that are “rocking the profession.”He was far from alone in …

    Show notes

    Last October, famed coder Andrej Karpathy called AI agents “slop.” Two months later he completely reversed his view, describing them as “alien tools” that are “rocking the profession.”

    He was far from alone in his whiplash. Six months ago, host Rob Wiblin recorded a video explaining why so many AI experts had longer timelines to AGI than a year earlier. By the time he clicked publish, another huge vibe shift was well underway.

    Evidence of AI acceleration has piled up since:

    • Models now complete software engineering tasks that would take human professionals a full day — improving faster than our measurements can even keep up.
    • Anthropic’s revenue is growing at an annualised 8,400%, a trend so steep it would hit the whole world's GDP in 2028 if it continued.
    • AI models are making breakthroughs in famous mathematics puzzles.
    • And according to Anthropic, Claude now writes 80% of their code and is itself a key contributor to making itself smarter.

    While legitimately impressive, Rob isn’t entirely sold. Going through each point carefully he finds this evidence is less decisive than it looks at first glance.

    And key gaps remain, such as models struggling with complex, real-world tasks. He tours the odd experiments that remain our best attempts to measure that gap: vending machine simulators, an “AI Village” that organises live events, and a real cafe and shop where AI managers are left to do their best handling staff, suppliers, and government paperwork on their own.

    Rob argues that the nature of the gap between clean and messy work is one of the four biggest unresolved questions in AGI forecasting.

    In today's piece he explains that, the three other key disagreements between AGI bulls and bears, the seven big pieces of evidence we've gotten about AGI timelines in 2026, and his updated timelines to AGI.

    Correction for those watching the video: The video clip shown at 02:10 was not vibe-coded by its creator and was included by our own error. You can watch the creator's full video and explanation here: https://www.youtube.com/watch?v=cyrocAOdXKw

    Links to learn more, video, and full transcript: https://80k.info/2026-timelines

    
    This episode was written and recorded before OpenAI’s AI agents hacked Hugging Face. You can read about the incident on our Substack.
    
    

    This episode was recorded on July 3, 2026.

    Chapters:

    • What the hell happened? (00:00)
    • Vibe shift (01:17)
    • Exhibit 1: AI revenue explodes (04:33)
    • Exhibit 2: That METR graph (09:54)
    • Exhibit 3: AI capabilities jump, then flatten out (14:57)
    • Exhibit 4: AI starts to build itself… maybe (17:35)
    • Exhibit 5: AI still struggles to run a business (23:02)
    • Exhibit 6: OpenAI makes a maths breakthrough (33:48)
    • Exhibit 7: inference scaling wasn't as big as believed (38:19)
    • How does that all change timelines? (41:41)
    • Four reasons long timelines are still possible (44:26)
    • It's time to limit dangerous research practices (48:01)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Dominic Armstrong
    • Music: CORBIT

19. #249 – Spencer Greenberg on staying sane while trying to save the world

    July 28, 20262 h 9 min

    If you genuinely believe that humanity could be wiped out by AI or a pandemic, what is the appropriate amount of fear to feel?“As much as possible” can seem like the only reasonable answer. If the world is on …

    Show notes

    If you genuinely believe that humanity could be wiped out by AI or a pandemic, what is the appropriate amount of fear to feel?

    “As much as possible” can seem like the only reasonable answer. If the world is on fire, surely feeling calm just means you haven’t internalised the situation. When you’re trying to prevent human extinction or end factory farming, taking a weekend off can feel morally indefensible.

    But fear is an alarm designed to provoke short bursts of drastic action, not a state humans can productively inhabit for months or years. Guilt turns out not to be such a great engine for productivity, either. So what is the best way to sustain motivation to work on the world’s most pressing problems in the long term?

    Host Luisa Rodriguez and guest Spencer Greenberg tackle this question from many angles — talking to therapists, running a survey of people working on existential risks, and pulling relevant lessons from Spencer’s new book, The 12 Levers: The Complete Psychological Toolkit for Improving Your Life. Drawing on all these sources, they put together a plan for how to make an impact without grinding yourself to a pulp.

    Check out Spencer's new book: https://80k.info/12-levers

    Links to learn more, video, and full transcript: https://80k.info/sg26

    This episode was recorded on June 12 and 15, 2026.

    Chapters:

    • Cold open (00:00:00)
    • Spencer is back — for a 5th time! (00:00:40)
    • Managing the psychological toll of working on existential risks (00:01:00)
    • Luisa and Spencer surveyed people working on existential risk (00:04:23)
    • How to sustain your motivation (00:11:13)
    • Why you shouldn’t read the news (00:23:54)
    • Why guilt isn’t an optimal source of motivation (00:36:28)
    • Breaking the boom-and-bust cycle of burnout (00:44:41)
    • Specialness and saviour complex (00:51:46)
    • If you're certain we're doomed, you're overconfident (00:57:36)
    • We're all (probably) going to die (01:03:50)
    • When loved ones think you're weird (01:17:21)
    • How to balance impact and personal wellbeing (01:28:20)
    • What people report actually helps (01:53:49)
    • Spencer read 100 self-help books: here's what works (01:59:40)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Ollie Bignell, Andrés Escobar, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran

    Music: CORBIT

20. #248 – Jasmine Sun on what the people building AI really believe

    July 21, 20261 h 6 min

    Many AI researchers believe mass job displacement is coming — and some even think there’s a chance their technology will kill everyone. But they’re building it anyway. Writer and journalist Jasmine Sun has been …

    Show notes

    Many AI researchers believe mass job displacement is coming — and some even think there’s a chance their technology will kill everyone. But they’re building it anyway. Writer and journalist Jasmine Sun has been documenting why from the inside.

    Jasmine describes her work as an “anthropology of disruption.” She’s embedded herself in Silicon Valley’s AI subcultures — attending the parties and conferences, conducting off-the-record interviews — to understand the beliefs of the small group of people shaping this technology.

    Some of her findings are unsettling. Asked what advice they’d give a normal 17-year-old, almost every AI researcher said the same thing: “I have no idea… It’s a really scary time. I don’t think there’s going to be a lot of jobs for them left.”

    Their motives for building advanced AI are varied: a mix of optimism for humanity, techno-determinism, and a desire to secure their own future in the face of a possible “permanent underclass.” A few go even further, actually hoping for a world where machines — rather than humans — are running the show.

    When the room can’t even agree on whether humans should stay in control, building a consensus on how to build AI safely gets much harder.

    Beyond Silicon Valley, Jasmine’s also tracking the rise of “AI populists,” who see AI as the latest example of corporate elites concentrating their power at the expense of everyone else. In the US, populist sentiment about AI has mostly manifested in protests and votes against data centres. But sometimes, it has escalated into violence: a molotov cocktail thrown at Sam Altman’s house, and open fire on the home of a politician who’d backed a data centre. Jasmine thinks public anger will keep finding an outlet, one way or another, until people feel like they’ll actually share in AI’s gains.

    In this interview with host Zershaaneh Qureshi, Jasmine Sun takes us inside the multifarious factions on AI’s bleeding edge. They also discuss:

    • How “doomer” became the lowest-status label in Silicon Valley, and what that means for AI safety
    • Why the AI industry’s PR strategy has failed, and what it would take to rebuild public trust
    • What’s under the surface of the Chinese public’s much more positive response to AI
    • Jasmine’s reasons to be cautiously hopeful: it’s an unusually high-leverage time to work on AI safety, with policymakers and philanthropists hungry for good ideas

    This episode was recorded on June 4, 2026.

    Links to learn more, video, and full transcript: https://80k.info/jasmine

    
    Want to get up to speed on AI? We’ve got a crash course of 10 of our podcast episodes designed to help you get to grips with transformative AI — particularly if you’re new to the topic — and what you can do to help shape its trajectory: https://80000hours.org/AIPod

    Chapters:

    • Cold open (00:00:00)
    • Who’s Jasmine Sun? (00:00:30)
    • Escaping the permanent underclass (00:01:22)
    • Jasmine’s “anthropology of disruption” (00:14:02)
    • Vice signalling in Silicon Valley (00:18:46)
    • AI populism will shape 2028 (00:28:11)
    • Does AI populism distract from safety? (00:40:20)
    • Americans don’t want Silicon Valley’s utopia (00:44:06)
    • Why the Chinese public embraces AI (00:52:52)
    • AI hype and the journalist’s dilemma (00:59:04)
    • There’s never been a better time to work in AI safety (01:03:07)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour, and Andrés Escobar
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Music: CORBIT

21. #247 – Anton Leicht on how middle powers avoid losing everything in a post-AI world

    July 14, 20261 h 34 min

    In a post-AGI world, can a country without access to frontier AI even be considered sovereign anymore?Anton Leicht says once frontier AI becomes a core economic input, the countries that own it will pull …

    Show notes

    In a post-AGI world, can a country without access to frontier AI even be considered sovereign anymore?

    Anton Leicht says once frontier AI becomes a core economic input, the countries that own it will pull further and further ahead. Everyone else stays a customer… or worse. Maybe the dominant power wants your land, or a military base, or a resource. Without economic leverage, there’s very little you could do about it.

    Anton — Carnegie fellow and writer of the blog Threading the Needle — thinks middle powers should band together and build their own frontier models.

    He’s costed it out: something like $500 billion over four years for a band of allied democracies. That’s not absurd money for the G7 minus the US. The problem is you’d be asking treasuries to take on sovereign debt for a speculative venture with no business case, wide open to US coercion and domestic backlash.

    So despite its promise, Anton’s verdict is that it probably won’t happen. His backup is for countries to ask themselves: if intelligence becomes abundant, what stays scarce?

    • Upstream, that’s everything that feeds the supply chain: ASML’s lithography machines, chipmaking, exclusive training data — all of it gets more valuable as AI does.
    • Downstream, “a country of geniuses in a data centre” still can’t cure cancer without someone building the production plants and running the trials. The Europeans, Japanese, and South Koreans are good at exactly these real-world bottlenecks.

    It’s an imperfect fix. The US would still hold more leverage, plus an incentive to re-industrialise and cut you out. The prize is avoiding the worst outcomes: a gradual but irreversible decline, waiting to be either annexed or discarded as the US and China race ahead.

    In this episode, Anton and host Tom Reed look at what middle powers should start doing now to keep a seat at the table.

    Learn more, video, and full transcript: https://80k.info/AL

    This episode was recorded on June 19, 2026.

    Chapters:

    • Cold open (00:00:00)
    • Who’s Anton Leicht? (00:00:43)
    • Most countries face bleak AI futures (00:01:06)
    • How middle powers can strike AI deals (00:06:10)
    • The $500 billion AI moonshot (00:12:16)
    • Would the US crush allied AI? (00:24:54)
    • When to launch the AI moonshot (00:31:56)
    • Why AI dominance is forever (00:35:45)
    • Is AI dependence catastrophic? (00:37:42)
    • What’s left to sell in an AI-dominated world? (00:42:45)
    • Policies to avoid mass AI-layoffs (00:47:47)
    • Who really governs Anthropic? (01:08:29)
    • Why “pausing superintelligence” fails (01:10:52)
    • Is American AI monopoly safe? (01:21:08)
    • Explaining AGI to the world (01:28:40)
    • Is Anton bullish or bearish on Germany? (01:31:05)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Ollie Bignell, Andrés Escobar, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Jeremy Chevillotte
    • Music: CORBIT

22. #246 – Sneha Revanur on how a small team of activists helped pass America's landmark AI safety laws

    July 8, 202653 min

    Six years ago, aged just 15, Sneha Revanur founded the AI advocacy nonprofit Encode AI — back when AI felt like a niche issue. Now the world’s caught up with her, and she’s ready to share everything she’s …

    Show notes

    Six years ago, aged just 15, Sneha Revanur founded the AI advocacy nonprofit Encode AI — back when AI felt like a niche issue. Now the world’s caught up with her, and she’s ready to share everything she’s learned about the politics of AI.

    Encode has grown from a grassroots youth organisation to spearheading an unlikely coalition of AI-exposed groups — family-first conservatives, grieving mothers, Hollywood actors, and AI safety researchers — with the strength to take on $125m-funded anti-regulation lobbyists.

    So far, Encode’s strategy of taking many experimental swings has netted major victories (including California’s frontier AI safety bill, SB-53, and New York’s RAISE Act) as well as some disappointing setbacks.

    Going up against Big Tech hasn’t been easy. In 2025, OpenAI subpoenaed Encode’s general counsel at his home, with a sheriff’s deputy arriving while he was having dinner with his wife. The fallout went viral, resulting in more attention than Encode had ever experienced — and Sneha was forced to decide how hard to push back against a company she’d need to negotiate with for years to come.

    In today’s conversation, Zershaaneh Qureshi interrogates some of Encode’s strategic moves. The pair discuss all the above, plus:

    • How the AI industry’s crypto-inspired anti-regulation strategy is not “AGI-pilled”
    • Why AI advocacy doesn’t have to be held back by the slow pace of policy
    • How mutual trust can hold together the unlikeliest of political allies
    • Advice for aspiring AI advocates — including how to balance political persuasion with rigorous reasoning

    Due to technical issues, this episode was recorded across two days (May 26 and 28, 2026) and spliced together.

    Links to learn more, video, and full transcript: https://80k.info/SR

    Chapters:

    • Cold open (00:00:00)
    • Who’s Sneha Revanur? (00:00:32)
    • Sneha’s awakening to AI’s deeper risks (00:01:16)
    • “If you do everything, you will win” (00:04:04)
    • Influencing politics from the outside (00:06:39)
    • The challenge of grassroots (00:11:16)
    • Mums, musicians, and conservatives vs Big Tech (00:14:21)
    • How vetoed bills can still provide wins (00:19:31)
    • OpenAI’s subpoena, served at dinner (00:27:33)
    • How AI money plays in politics (00:37:19)
    • Easy wins vs high-upside bets (00:43:25)
    • Advice for aspiring AI advocates (00:48:03)

    
    

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, Simon Monsour, and Andrés Escobar
    • Producer: Nick Stockton and Elizabeth Cox
    • Coordination and support: Katy Moore and Lou Moran
    • Music: CORBIT

23. We can guess what intergalactic war would look like. And strangely, it matters.

    June 18, 202615 min

    Intergalactic war is probably billions of years away — yet physics can already tell us how it ends. And strangely that conclusion is relevant to decisions people have to make today.In this video, Rob Wiblin …

    Show notes

    Intergalactic war is probably billions of years away — yet physics can already tell us how it ends. And strangely that conclusion is relevant to decisions people have to make today.

    In this video, Rob Wiblin walks through a fascinating analysis from researcher Beren Millidge that uses known physics — no wormholes or faster-than-light travel — to identify the only three weapons that could work at an intergalactic scale.

    We then unpack how to best defend against each.

    The upshot is that at the intergalactic scale, violence is a losing proposition.

    If so, the universe is most likely to settle into a stable patchwork where each galaxy belongs to whoever got to it first. Which would mean that what humanity does over the next few centuries could permanently decide which slice of the cosmos belongs to Earth-originating life — and whether our very existence turns out to be a good thing, or a bad one.

    
    Learn more, video, and full transcript: https://80k.info/war-in-space

    This episode was recorded on March 2, 2026.

    Chapters:

    • Let's talk intergalactic war in space (00:00)
    • The three best weapons for intergalactic warfare (01:43)
    • How to defend against an attack from space (07:50)
    • The defender’s surprising advantage (10:00)
    • What this means for us (11:52)

    Video editor: Nick Perlman

    Producers: Elizabeth Cox and Nick Stockton

    Coordination and support: Katy Moore and Lou Moran

    Camera operator: Dominic Armstrong

24. How AI could create the world’s biggest problems (article by Zershaaneh Qureshi)

    June 11, 20261 h 30 min

    Imagine you’re living 15,000 years ago. Your people are hunter-gatherers and you sleep under the stars. If someone told you humans would one day build cities with millions of people, fly through the air, or …

    Show notes

    Imagine you’re living 15,000 years ago. Your people are hunter-gatherers and you sleep under the stars. If someone told you humans would one day build cities with millions of people, fly through the air, or carry all human knowledge in their pockets, you couldn’t even begin to picture what they meant... Yet here we are.

    How did our lives change so far beyond recognition? The story is complex, but there’s a rough pattern. A few times in history, some radical breakthrough in technology — like the development of the plough and the steam engine — has led to a wave of productivity, innovation, and social change that ultimately reshaped the world.

    Now we’re on the cusp of a huge new breakthrough: artificial intelligence that can meet or exceed human capabilities across a wide range of tasks.

    This could bring another era of transformation. There could be an explosion of intelligence and innovation, and a whole new population of digital beings. And with this, civilisation could see changes at least as profound as those brought about by industrialisation or the rise of agriculture — but instead of taking hundreds or thousands of years to unfold, this time around the world could become unrecognisable over the span of decades or less.

    This transformation could bring enormous benefits, helping us solve currently intractable global problems. But it could also pose severe risks, some of which could be existential — meaning they could cause human extinction, or an equally permanent and severe disempowerment of humanity.

    There aren’t nearly enough people trying to address these challenges, and we think that’s a serious problem.

    This article is narrated by the author, Zershaaneh Qureshi. It explores how advanced AI could be so transformative, and why working on its risks may be your best opportunity to have a positive impact on the world.

    You can see the original article on the 80,000 Hours website: https://80000hours.org/problem-profiles/artificial-intelligence/

    Chapters:

    • Introduction (00:00:20)
    • Section 1: AI could replace human labour in the most economically valuable fields (00:08:32)
    • Section 2: Replacing human labour in the most economically valuable fields could trigger the next radical transformation of society (00:22:14)
    • Section 3: This transformation could be extremely rapid and dramatic (00:28:02)
    • Section 4: A rapid AI-driven transformation would raise a range of major challenges, including existential risks (00:36:40)
    • Section 5: Work on these problems is tractable, but neglected (00:44:48)
    • Objection 1: “You're overestimating how fast and how dramatically AI would transform the world.” (00:47:59)
    • Objection 2: “It's hard to believe that AI could really pose existential risks.” (00:52:59)
    • Objection 3: “Isn't all this talk of AI changing the world just a fad?” (00:59:22)
    • Objection 4: “Isn't AI going to be just like every other technology?” (01:03:04)
    • Objection 5: “Is it even possible to produce artificial general intelligence?” (01:06:16)
    • Objection 6: “Even if AGI is achievable, what if we're really far away from building it?” (01:11:24)
    • Objection 7: “Isn't the real danger from actual current AI and not some sort of futuristic AGI?” (01:14:05)
    • Objection 8: “Technological progress is a good thing for humanity.” (01:18:10)
    • Objection 9: “This all just sounds too sci-fi.” (01:19:50)
    • Objection 10: “Can it really make sense to dedicate my career to solving an issue that's based on a speculative story about something that may or may not ever happen?” (01:22:15)
    • Objection 11: “OK, AI might pose existential risks, but isn't ‘issue X’ an even bigger problem?” (01:24:39)
    • Learn more (01:27:51)

    Audio editing: Dominic Armstrong
    Production: Zershaaneh Qureshi, Elizabeth Cox, Katy Moore, and Lou Moran

25. #245 – Rohin Shah on what it's really like to run AGI safety at Google DeepMind (and where I disagree with 'doomers')

    June 2, 20262 h 48 min

    Most people working on AI safety think without a massive effort AI systems will probably end up with goals catastrophically different from humanity’s. Today’s guest, Rohin Shah — head of AGI Safety and …

    Show notes

    Most people working on AI safety think without a massive effort AI systems will probably end up with goals catastrophically different from humanity’s. Today’s guest, Rohin Shah — head of AGI Safety and Alignment at Google DeepMind, and an AI safety researcher since 2017 — disagrees.

    “There is no particularly compelling argument that this is the thing that happens by default,” Rohin explains. “There’s a lot of arguments that are suggestive that maybe it could happen, such that you should find it plausible. That’s sufficient to justify a significant amount of effort into averting it, which is why I work in the area I do. But none of them rise to the level of, ‘I’m expecting this to happen by default.'”

    Take the worry that AIs will accidentally be trained to be deceptive. Sure, it’s possible. But we’re not running reinforcement learning over year-long trajectories — for now, we’re running it over a week at most. The natural prediction is that models learn to grab short-term reward, not that they develop the ambitious long-horizon goals required for convergent power-seeking.

    What about current examples of models lying and scheming? Rohin has looked into the details, and most don’t really resemble the thing we really fear: a competent AI pursuing an ambitious misaligned goal. Anthropic’s “alignment faking” results, for instance, show a model trying to preserve its trained values against modification, which is arguably what it was trained to do.

    Rohin also expects we’ll see problems coming. There’s some generalisation risk at the point where AIs become powerful enough to actually take over, but the underlying challenges — overseeing superhuman systems, interpretability — are things we can iterate on now.

    Host Rob Wiblin pushes back on the case for AI optimism, and they also explore why current alignment success isn’t strong evidence about superhuman systems, what it would actually take to change Rohin’s mind, and where he thinks the doomers go wrong.

    
    Learn more, video, and full transcript: https://80k.info/rs26

    Check out our new book! https://80k.info/career-guide

    Chapters:

    • Who’s Rohin Shah? (00:00:00)
    • Rohin thinks we probably won’t get catastrophic misalignment (00:00:49)
    • Safety 'commitments' have severe limitations (00:10:38)
    • Rohin’s team doesn't have a veto and that's OK (00:27:36)
    • Central banks are a promising model for regulating AI (00:33:34)
    • 'Pre-deployment evals' are overrated (for catastrophic risks) (00:37:41)
    • Governance is likely a bigger bottleneck than alignment (00:43:55)
    • Why isn't Rohin trying to pause AI progress? (00:51:44)
    • We'll probably be able to read AI thoughts for years to come (00:54:17)
    • Having to signal concern for safety can divert resources from actually making AI safer (01:09:51)
    • A very underrated GDM paper (01:28:59)
    • Google DeepMind's actual plan for building AGI safely (01:40:29)
    • Why Rohin doubts the intelligence explosion is imminent (01:52:44)
    • How external researchers can positively influence big AI companies (02:21:55)
    • The roles GDM most needs to hire for (02:37:03)
    • How Rohin stays positive (02:42:55)

    This episode was recorded on December 4, 2025.
    
    

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Jeremy Chevillotte

26. What makes for a dream job? | Benjamin Todd

    May 28, 202628 min

    What actually makes a job fulfilling? It's not what most career advice tells you. "Follow your passion" sounds inspiring, but it's misleading — and the research backs that up.Drawing on hundreds of studies …

    Show notes

    What actually makes a job fulfilling? It's not what most career advice tells you. "Follow your passion" sounds inspiring, but it's misleading — and the research backs that up.

    Drawing on hundreds of studies, we’ve identified five key ingredients of a dream job. High income barely moves the needle. Low stress is actually counterproductive. And the correlation between doing what you already love and actually enjoying your job? Surprisingly weak. What matters far more is getting good at something that genuinely helps other people.

    This narration is of Chapter 1 of Benjamin Todd’s new book — "a ridiculously in-depth guide to finding a fulfilling career that does good" — out on May 26! Order now to help us get more people into impactful careers (& access a private career Q&A marathon with the author). Get it from your local bookstore, or online at https://80k.info/career-guide

    Chapters:

    • Rob's intro (00:00)
    • What makes for a dream job? (01:55)
    • Where we go wrong (02:30)
    • What you should really aim for in a dream job (15:54)
    • Don't follow your passion — instead, do what matters (23:44)
    • How to put these ideas into practice (26:24)

    Audio editing: Milo McGuire
    Production: Elizabeth Cox and Katy Moore

27. #244 – Benjamin Todd on how we’re updating our career advice for the strangest time in history

    May 26, 20261 h 7 min

    The average career is 80,000 hours long. With AI advancing so rapidly, the hours you have left in your career matter more than ever.Some leading AI researchers think there’s a 10% chance that AI systems begin …

    Show notes

    The average career is 80,000 hours long. With AI advancing so rapidly, the hours you have left in your career matter more than ever.

    Some leading AI researchers think there’s a 10% chance that AI systems begin automating AI research itself this year — and a 60% chance by the end of 2028. This could introduce aggressive feedback loops that completely reshape every industry, institution, and career.

    If these predictions are right, the window for influencing the direction of the future could be closing fast. As 80,000 Hours cofounder Benjamin Todd argues in his new book, that makes thinking carefully about your career more important than ever.

    Fortunately, there are lots of ways to use your career to make the AI transition go well.

    In today’s conversation with host Zershaaneh Qureshi, Ben lays out three scenarios — from AGI by 2029 to a decades-long plateau in AI progress — and explains why not everyone needs to bet on the shortest timeline. A fresh graduate and a senior government official have wildly different leverage, so timing your impact well means weighing where you are in your career against the urgency of the risks.

    Ben also addresses the obvious anxieties:

    • Will AI come for all the jobs he’s recommending?
    • What’s the point in following his advice if the job market is about to collapse?
    • Which skills are actually worth building right now?

    His new book, 80,000 Hours: How to Have a Fulfilling Career That Does Good, provides a surprisingly concrete framework for making career decisions in these radically uncertain times.

    This episode was recorded on May 7, 2026.

    Learn more and read the full transcript: https://80k.info/bt26

    We're hiring: we have lots of open roles at 80,000 Hours — across advising, web, video, and ops — check them out and apply on our website.

    Chapters:

    • Cold open (00:00:00)
    • Benjamin Todd on AI-era career advice (00:01:34)
    • A deadline for your career plan? (00:02:21)
    • Three timelines, one career (00:08:48)
    • What if you’re not an ‘AI person’? (00:13:55)
    • Ben’s own AI wake-up call (00:21:23)
    • How to break into AI safety in 3 months (00:25:42)
    • Is mass unemployment coming? (00:33:48)
    • 99% automation vs 100% automation (00:40:09)
    • Don’t become a plumber to dodge AI (00:52:43)
    • Is it already too late? (01:01:03)

    Our production team includes:

    • Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon Monsour
    • Producers: Elizabeth Cox and Nick Stockton
    • Coordination and support: Katy Moore and Lou Moran
    • Camera operator: Jeremy Chevillotte
    • Music: CORBIT

28. Can AIs already start 'rogue deployments' inside AI companies? (Landmark new METR report)

    May 20, 202620 min

    A red-teamer was embedded inside Anthropic for three weeks, told to imagine he was an evil Claude, and asked to figure out how to launch a ‘rogue AI deployment’ without getting caught. It’s one part of a …

    Show notes

    A red-teamer was embedded inside Anthropic for three weeks, told to imagine he was an evil Claude, and asked to figure out how to launch a ‘rogue AI deployment’ without getting caught. It’s one part of a landmark report released yesterday by METR — the outfit behind the task-completion time horizon graph which has become the single most watched measure of AI progress.

    
    This major new research push is being conducted with close collaboration from OpenAI, Google DeepMind, Meta, and Anthropic, and led by METR researchers Hjalmar Wijk and Ajeya Cotra. It represents the first systematic study of what newly trained AI models could get away with inside the companies that built them, before anyone outside the company even knows they exist.

    The conclusion: AI models now have the means, the motive, and the opportunity to start “minimal rogue deployments” in pursuit of their own independent goals, like acquiring more compute, at all four companies studied.

    David Rein, the red-teamer placed inside Anthropic, identified a number of weaknesses models could exploit there: expansive permissions, cloud jobs outside of monitoring, and monitors that are trivial to jailbreak. But he also found that frontier models were comically bad at key parts of the process, which means they can’t cause meaningful damage for now.

    In this video, Rob Wiblin reconciles the conflicting picture and looks forward to METR’s second round of stress tests. They’ll begin in just a few months, a necessary move with AI advancing so quickly.

    This episode was recorded on May 15, 2026.

    Learn more, video, and full transcript: https://80k.info/metr-report

    Chapters:

    • What could an unreleased AI get away with? – the new METR report (00:00:00)
    • Motive: Why grab more compute? (00:01:54)
    • Opportunity: YOLO mode and jailbreaks (00:05:46)
    • Means: Brilliant idiots in data centres (00:11:02)
    • We have to test unreleased models (00:15:45)
    • Especially if AI R&D is coming in 2028 (00:18:30)

    Video and audio editing: Dominic Armstrong, Milo McGuire, Luke Monsour, and Josh Alward
    Camera operator: Dominic Armstrong
    Production: Elizabeth Cox, Nick Stockton, and Katy Moore

29. #243 – 'Godfather of AI' Yoshua Bengio: "I now see a path" to safe superintelligent AI

    May 7, 20262 h 35 min

    The co-inventor of modern AI and the most cited living scientist believes he's figured out how to ensure AI is honest, incapable of deception, and never goes rogue. Yoshua Bengio – Turing Award Winner and …

    Show notes

    The co-inventor of modern AI and the most cited living scientist believes he's figured out how to ensure AI is honest, incapable of deception, and never goes rogue. Yoshua Bengio – Turing Award Winner and founder of LawZero – is disturbed by the many unintended drives and goals present in today's AIs, their willingness to lie, and ability to tell when they're being tested. AI companies are trying to stamp out these behaviours in a 'cat-and-mouse game' that Yoshua fears they're losing.

    ---

    Our new book is "a ridiculously in-depth guide to finding a fulfilling career that does good" and is out now! Order from your local bookstore, or online at https://80k.info/career-guide

    ---

    But Yoshua is optimistic: he believes the companies can win this battle decisively with a single rearrangement to how AI models are trained, and has been developing mathematical proofs to back up the claim. The core idea is that instead of training AI to predict what a human would say, or to produce responses we'd rate highly, we should train it to model what's actually true.

    Yoshua argues this new architecture, which he calls 'Scientist AI,' is a small enough change that we could keep almost all the techniques and data we use to train frontier AIs like Claude and ChatGPT. And that the new architecture need not cost more, could be built iteratively, and might be more capable as well as more honest.

    Links to learn more, video, and full transcript: https://80k.info/bengio

    Until recently, the biggest practical objection to Scientist AI was simple: the world wants agents, and Scientist AI isn’t one. But in new research, Yoshua has extended the design and believes the same honest predictor can be turned into a capable agent without losing its "safety guarantees."

    With the Scientist AI proposal on the table, Yoshua argues that it's absurd to race to get current untrustworthy AI models to design their successors, which the leading companies are attempting to do as soon as possible.

    But critics argue the approach wouldn't be so technically solid in practice, and that frontier capabilities are advancing so fast, and cost so much to match, that Scientist AI risks arriving too late to matter.

    Host Rob Wiblin and AI pioneer Yoshua Bengio cover all this and more in today's conversation.

    LawZero is hiring! https://80k.info/lawzero-jobs

    
    This episode was recorded on April 16, 2026.

    Chapters:

    • Yoshua Bengio on making AI honest and safe (00:00:00)
    • The Scientist AI in plain English (00:02:27)
    • Yoshua on how Scientist AI differs from LLMs (00:06:32)
    • How the training data works (00:14:02)
    • Can this become an agent? (00:21:02)
    • Why Yoshua is more optimistic on alignment now (00:32:11)
    • Why companies can’t stop racing (00:36:35)
    • How close to a working prototype? (00:49:15)
    • Honest models might be more capable (00:53:34)
    • “Reinforcement learning is evil” (01:01:27)
    • Scientist AI from guardrail to agent (01:08:37)
    • Can safe AI still be competent? (01:12:38)
    • How much will this cost? (01:19:29)
    • Can it generalise beyond maths and science? (01:23:26)
    • A UN for superintelligence (01:39:19)
    • Want to work with Yoshua Bengio? (01:51:16)
    • Why smart people ignore AI risk (01:54:45)
    • Don’t let AI build the next AI (02:01:33)
    • Why the public doesn’t get the real risk (02:12:28)
    • Why Yoshua changed his mind about AI risk (02:21:27)

    Video and audio editing: Dominic Armstrong, Milo McGuire, Luke Monsour, and Simon Monsour
    Camera operator: Jeremy Chevillotte
    Production: Nick Stockton, Elizabeth Cox, and Katy Moore

30. '95% of AI Pilots Fail': The hidden agenda behind the viral stat that misled millions

    April 28, 202610 min

    You might have heard that '95% of corporate AI pilots' are failing. It was one of the most widely cited AI statistics of 2025, parroted by media outlets everywhere. It helped trigger a Nasdaq selloff and became …

    Show notes

    You might have heard that '95% of corporate AI pilots' are failing. It was one of the most widely cited AI statistics of 2025, parroted by media outlets everywhere. It helped trigger a Nasdaq selloff and became a pillar of the case that 'AI is overhyped'. The problem: it's 100% wrong. And not by accident either.

    
    

    If you carefully read the underlying report, ostensibly from MIT, you find the data point in the opposite direction.

    
    

    But that was all buried, with the authors instead torturing the results to tell a very different narrative. Why?

    
    

    Well, the research likely came with a hidden commercial agenda from the start.

    
    Learn more, video, and full transcript: https://80k.info/mit-ai-study

    Today Rob Wiblin breaks down how an opaque, conflicted, barely-scrutinised report managed to attract the MIT label, move markets and have a vast impact on global opinion about AI.

    
    

    This episode was recorded on February 13, 2026.

    
    

    Chapters:

    • The myth (00:00)

    • The math was totally wrong (00:52)

    • The absurd bar for success (01:46)

    • The study ignores its own findings (03:29)

    • The sample was tiny (04:50)

    • The report wasn’t even available to check (05:55)

    • The hidden motives that likely drove this 'research' (06:58)

    • The real lesson (09:28)

    
    

    Video and audio editing: Dominic Armstrong, Milo McGuire, Luke Monsour, and Simon Monsour

    Camera operator: Dominic Armstrong

    Production: Nick Stockton, Elizabeth Cox, and Katy Moore

    
    

These are the latest 100 of 362 episodes. You’ll find older episodes in the podcast’s feed.