How to go from zero to world-class in AI safety in 1-2 years — no fellowship needed
austinc3301 · x · 2026-11-08
Responding to the observation that people can go from zero to world-leading on important topics in 1-2 years, the author clarifies this isn't about joining MATS — there's plenty of work outside AI safety fellowships.
A quick brainstorm of concrete entry points:
- Replicate important AI safety papers and results
- Synthesize findings from multiple papers on one topic, reconcile or flag contradictions, and suggest what work could resolve them
- Read and summarize companies' risk reports and AI safety research
- Pick important claims in risk reports and write explanations or stress-test them
- Write and maintain up-to-date summaries of AI safety research areas
More from AGI Musings
- ChrisGPT: We could automate most work in 10 years, yet fear-driven bans dominate — ChrisGPT · 2026-11-08
- "All the bad stuff in society is going to go away": Brad Jacobs' AI optimism mocked as INFINITE FUN SPACE — generativist · 2026-11-08
- Researcher warns AI being right is the bigger danger: we may lose the ability to tell truth from falsehood — ValerioCapraro · 2026-11-08
- METR Assesses Anthropic's AI R&D Acceleration and Its Risk Conclusions — Matthew Berman · 2026-11-08
- KOL: a 500-year-old human might be a completely different person — how radical is immortality? — VraserX · 2026-11-08
- Richard Ngo channels Nietzsche on living in an age when civilization's foundations shake — RichardMCNgo · 2026-11-08