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)
Why Rohin thinks we won’t get catastrophic misalignment (00:00:49)
The limitations of safety and alignment commitments (00:10:38)
Does Rohin’s team have veto power at Google DeepMind? (00:27:36)
Central banks as a roadmap for regulating AI (00:33:34)
How useful are pre-deployment evaluations of models? (00:37:41)
Governance might be a bigger bottleneck than alignment (00:43:55)
Why not just pause AI progress? (00:51:44)
How much longer will we be able to read AIs’ thoughts? (00:54:17)
Sometimes, having to signal concern for safety diverts resources from actually making AI safer (01:09:51)
Underrated GDM paper: Training away hidden reward hacks (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)
Advice for external researchers who want to impact big AI companies (02:21:55)
The most in-demand roles at GDM (02:37:03)
How Rohin maintains his positivity (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

What it's really like to run AGI safety at Google DeepMind (and where I disagree with 'doomers') | Rohin Shah

80,000 Hours Podcast

What it's really like to run AGI safety at Google DeepMind (and where I disagree with 'doomers') | Rohin Shah