Chris Olah on what the hell is going on inside neural networks
Humanity made these amazing and ever-improving tools. So how do our creations work? In short: we don’t know.
80000hours.orgHere’s the latest I can share about Chris Olah based on recent public reporting.
Key developments
- Chris Olah remains a prominent figure in AI interpretability, with ongoing involvement in industry research and public discourse about mechanistic interpretability and safety in large language models.[2][3]
- In 2024 he was named to TIME’s 100 Most Influential People in AI, a recognition that has continued to frame his influence in the field.[7]
- He has continued to be associated with Anthropic, the AI safety and research lab, where he has been cited as a co-founder and a leading voice in interpretability research.[2]
- Public appearances and interviews into 2024–2025 highlight his ongoing communication on how neural networks work and the importance of interpretability work in production-scale AI systems.[3][4][5]
Representative sources you can check
- TIME’s profile on Chris Olah (2024) for context on why he was listed among AI influencers.[7]
- Anthropic and OpenAI-related research pages and interviews for updates on his role and the interpretability program he leads.[8][2]
- The 80,000 Hours podcast and related talks from 2024–2025 for his perspectives on mechanistic interpretability and career paths in AI research.[5][3]
If you’d like, I can pull the most current headlines and summarize any specific events (talks, papers, or interviews) from the latest weeks. I can also pull full articles or direct quotes if you want deeper detail.