Most AI safety conversations centre on alignment: ensuring AI systems share our values and goals. But despite progress, we’re unlikely to know we’ve solved the problem before the arrival of human-level and superhuman systems in as little as three years.

So some — including Buck Shlegeris, CEO of Redwood Research — are developing a backup plan to safely deploy models we fear are actively scheming to harm us: so-called “AI control.” While this may sound mad, given the reluctance of AI companies to delay deploying anything they train, not developing such techniques is probably even crazier.

These highlights are from episode #214 of The 80,000 Hours Podcast: Buck Shlegeris on controlling AI that wants to take over – so we can use it anyway, and include:

• What is AI control? (00:00:15)
• One way to catch AIs that are up to no good (00:07:00)
• What do we do once we catch a model trying to escape? (00:13:39)
• Team Human vs Team AI (00:18:24)
• If an AI escapes, is it likely to be able to beat humanity from there? (00:24:59)
• Is alignment still useful? (00:32:10)
• Could 10 safety-focused people in an AGI company do anything useful? (00:35:34)

These aren't necessarily the most important or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Ben Cordell, Milo McGuire, and Dominic Armstrong

Watch this episode on YouTube! https://youtu.be/fJssGodnCQg

Conor and Arden sit down with Matt in his farewell episode to discuss the law, their team retreat, his lessons learned from 80k, and the fate of the show.

The 20th century saw unprecedented change: nuclear weapons, satellites, the rise and fall of communism, third-wave feminism, the internet, postmodernism, game theory, genetic engineering, the Big Bang theory, quantum mechanics, birth control, and more. Now imagine all of it compressed into just 10 years.

That’s the future Will MacAskill — philosopher and researcher at the Forethought Centre for AI Strategy — argues we need to prepare for in his new paper “Preparing for the intelligence explosion.” Not in the distant future, but probably in three to seven years.

These highlights are from episode #213 of The 80,000 Hours Podcast: Will MacAskill on AI causing a “century in a decade” — and how we’re completely unprepared, and include:

• Rob's intro (00:00:00)
• A century of history crammed into a decade (00:00:17)
• What does a good future with AGI even look like? (00:04:48)
• AI takeover might happen anyway — should we rush to load in our values? (00:09:29)
• Lock-in is plausible where it never was before (00:14:40)
• ML researchers are feverishly working to destroy their own power (00:20:07)
• People distrust utopianism for good reason (00:24:30)
• Non-technological disruption (00:29:18)
• The 3 intelligence explosions (00:31:10)

These aren't necessarily the most important or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected]. 

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

Technology doesn’t force us to do anything — it merely opens doors. But military and economic competition pushes us through. That’s how Allan Dafoe — director of frontier safety and governance at Google DeepMind — explains one of the deepest patterns in technological history: once a powerful new capability becomes available, societies that adopt it tend to outcompete those that don’t. Those who resist too much can find themselves taken over or rendered irrelevant.

These highlights are from episode #212 of The 80,000 Hours Podcast: Allan Dafoe on why technology is unstoppable & how to shape AI development anyway, and include:

• Who's Allan Dafoe? (00:00:00)
• Astounding patterns in macrohistory (00:00:23)
• Are humans just along for the ride when it comes to technological progress? (00:03:58)
• Flavours of technological determinism (00:07:11)
• The super-cooperative AGI hypothesis and backdoors (00:12:50)
• Could having more cooperative AIs backfire? (00:19:16)
• The offence-defence balance (00:24:23)

These aren't necessarily the most important or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected]. 

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

Watch this episode on YouTube! https://youtu.be/IRRwHCK279E

Matt, Bella, and Huon sit down with Chi Nguyen to discuss cooperating with aliens, elections of future past, and Bad Billionaires pt. 2.

Check out: 

• Matt’s summer appearance on the BBC on funding for the arts
• Chi’s ECL Explainer (get in touch to support!)

Economist and editor of Works in Progress Sam Bowman isn’t content to just condemn the Not In My Back Yard (NIMBY) mentality behind rich countries' construction stagnation. He wants to actually get a tonne of stuff built, and by that standard the strategy of attacking ‘NIMBYs’ has been an abject failure. They are too politically powerful, and if you try to crush them, sooner or later they crush you.

So Sam lays out three alternative strategies in our full interview with him — including highlights like:

• Rich countries have a crisis of underconstruction (00:00:19)
• The UK builds shockingly little because of its planning permission system (00:04:57)
• Overcoming NIMBYism means fixing incentives (00:07:21)
• NIMBYs aren't wrong: they are often harmed by development (00:10:44)
• Street votes give existing residents a say (00:16:29)
• It's essential to define in advance who gets a say (00:24:37)
• Property tax distribution might be the most important policy you've never heard of (00:28:55)
• Using aesthetics to get buy-in for new construction (00:35:48)
• Locals actually really like having nuclear power plants nearby (00:44:14)
• It can be really useful to let old and new institutions coexist for a while (00:48:27)
• Ozempic and living in the decade that we conquered obesity (00:53:05)
• Northern latitudes still need nuclear power (00:55:30)

These highlights are from episode #211 of The 80,000 Hours Podcast: Sam Bowman on why housing still isn’t fixed and what would actually work. These aren't necessarily the most important, or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected]. (And you may have noticed this episode is longer than most of our highlights episodes — let us know if you liked that or not!)

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

We explored the cutting edge of wild animal welfare science our full interview with Cameron Meyer Shorb, executive director of Wild Animal Initiative, including highlights like:

• One concrete example of how we might improve wild animal welfare (00:00:16)
• How many wild animals are there, and which animals are they? (00:04:24)
• Why might wild animals be suffering? (00:08:40)
• The objection that we shouldn't meddle in nature because nature is good (00:12:25)
• Vaccines for wild animals (00:17:37)
• Gene drive technologies (00:20:50)
• Optimising for high-welfare landscapes (00:24:52)

These highlights are from episode #210 of The 80,000 Hours Podcast: Cameron Meyer Shorb on dismantling the myth that we can’t do anything to help wild animals. These aren't necessarily the most important, or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

Nonprofit legal expert Rose Chan Loui lays out the legal case and implications of OpenAI's attempt to shed its nonprofit parent. This episode is a selection of highlights from our full interview with Rose, including:

• How OpenAI carefully chose a complex nonprofit structure (00:00:26)
• The nonprofit board is out-resourced and in a tough spot (00:04:09)
• Is control of OpenAI 'priceless' to the nonprofit in pursuit of its mission? (00:06:47)
• Control of OpenAI is independently incredibly valuable and requires compensation (00:11:06)
• It's very important that the nonprofit gets cash and not just equity (00:16:06)
• How the nonprofit board can best play their hand (00:21:20)

These highlights are from episode #209 of The 80,000 Hours Podcast: Rose Chan Loui on OpenAI’s gambit to ditch its nonprofit. These aren't necessarily the most important, or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

Elizabeth Cox — founder of the independent production company Should We Studio — makes the case that storytelling can improve the world. This episode is a selection of highlights from our full interview with Elizabeth, including:

• Keiran’s intro (00:00:00)
• Empirical evidence of the impact of storytelling (00:00:16)
• The hits-based approach to storytelling (00:03:35)
• Debating the merits of thinking about target audiences (00:07:48)
• Ada vs other approaches to impact-focused storytelling (00:13:15)
• Why animation? (00:18:56)
• How long will humans stay relevant as creative writers, given AI advances? (00:22:40)

These highlights are from episode #208 of The 80,000 Hours Podcast: Elizabeth Cox on the case that TV shows, movies, and novels can improve the world. These aren't necessarily the most important, or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong

Charity founder Sarah Eustis-Guthrie has a candid conversation about her experience starting and running her maternal health charity, and ultimately making the difficult decision to shut down when the programme wasn’t as impactful as they expected. This episode is a selection of highlights from our full interview with Sarah:

• Luisa’s intro (00:00:00)
• What it's like to found a charity (00:00:14)
• Yellow flags and difficult calls (00:03:17)
• Disappointing results (00:06:28)
• The ups and downs of founding an organisation (00:08:37)
• Entrepreneurship and being willing to make risky bets (00:12:58)
• Why aren't more charities shutting down? (00:16:50)
• How to think about shutting down (00:19:39)

These highlights are from episode #207 of The 80,000 Hours Podcast: Sarah Eustis-Guthrie on why she shut down her charity, and why more founders should follow her lead. These aren't necessarily the most important, or even most entertaining parts of the interview — so if you enjoy this, we strongly recommend checking out the full episode!

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Simon Monsour, Milo McGuire, and Dominic Armstrong