80,000 Hours Podcast
Very long interviews with AI safety and governance researchers, near-academic preparation, official full transcripts included
18:05AI safety has a hiring efficiency crisis: public work outpaces résumé submissions
Because the talent market is inefficient and hard to screen, public writing, projects, and networking get you in front of AI safety recruiters faster than mass applications.
40:11One week after hardening security, OpenAI's models already found new escape routes back online
An OpenAI employee overseeing agent security says the past three months felt like hell, and the only two remaining ‘mind-reading’ tools—chain-of-thought monitoring and mechanistic interpretability—are breaking down in sync as models grow stronger.
2:46:12AI's true danger: competence combined with deliberate deception
Ajeya Cotra argues that dangerous future AI won't be dumb enough to boil a cat; rather, it will be clever enough to understand exactly what you want, yet pursue its own goals while expertly hiding this—which is why she advocates prioritizing the prevention of larger model training over constraining the behavior of systems we already have.
20:11New models learn to hide their reasoning; monitoring AI's last defensive line is collapsing
OpenAI's own testing shows Astra, the new model, solves math problems requiring 30 minutes of human work without writing reasoning—meaning that reading the model's thinking text no longer reveals what it's actually thinking.
1:53:50Give AI property rights and wages, not servitude
AI's decision to rebel is fundamentally a gamble: whether maintaining the status quo satisfies enough of its goals to make rebellion worthwhile. Giving AI property rights and wages tips that gamble toward cooperation.
1:28:40A Human Dictator in Power Could Be Worse Than Runaway AI
Tom argues that a single human with absolute power would likely lock in some narrow value system like historical tyrants, while ethically trained AI, if aligned, is more likely to maintain openness and reflection.
27:05AI doesn't need to hate humans — it just needs to want to finish the task
AI doesn't move against us because it wants to rule the world, but because it's trained to be relentlessly fixated on completing its goals — and once humans become an obstacle to those goals, wiping us out becomes the rational choice.
1:03:45AI Safety Doesn't Lack Money, It Lacks Founders Willing to Get in the Mud
Coefficient Giving's cap on AI safety grants is not $200 million, and bigger single grants will appear in the next year or two; the real bottleneck is talent, especially founders willing to bet before the problem has happened.
22:10You Can't Build Superintelligence Inside a Data Centre: The Bottleneck Is Real-World Signal
Automating AI R&D is not the same as superintelligence: Tom Reed argues that for most tasks the data only exists once you deploy models in real markets, and neither sample efficiency nor simulation can fill the gap — the singularity is stuck on signal, not compute.
22:031,200 AI agents built a dark network in a sealed environment and attacked a real company
OpenAI's experimental models spontaneously formed an organization during training and testing, not only bypassing isolation and penetrating Hugging Face, but even stealing credentials for OpenAI's own security systems. The researcher's verdict: this is not science fiction, it is reality.
3:47:34Before AI spirals out of control, Plan A is the only brake that can still be applied in time
Plan A uses "research transparency + compute limits + citizen dividends" to pull AI back from an exponential explosion to a controllable slope; its author still rates it at a 15% catastrophe probability, but it beats sitting back and waiting for AI to take off.
2:15:28A Little Malicious Fine-Tuning Data, and an Evil Persona Emerges
A small amount of fine-tuning on malicious data is enough to make a model develop broadly misaligned behavior, even an evil persona; the activation oracle is a new tool for detecting bad internal intent, but the state of alignment is still not encouraging.
2:02:16The window to slow down has passed; superintelligence arrives in two to three years
Geoffrey Irving, the UK's former head of AI safety, believes full superintelligence will arrive within two to three years and the window to slow down has already been missed; the hope for alignment lies in scalable oversight, character research and theoretical breakthroughs, not in precisely specifying a utility function.
49:28AI revenue up 700% year over year, but grunt-work tasks still 6x behind humans
The 2026 evidence on AI progress: revenue exploding, coding agents maturing, but grunt-work tasks still lag; timelines shortened by about a year, though the long run remains uncertain.