About this episode

Dwarkesh Patel interviews Ryan Greenblatt about the implications of AI automating AI research and the potential for rapid advancements in superintelligent systems.

Ryan Greenblatt is the Chief Scientist at Redwood Research, where he works on technical AI safety research. He's also lead author on the "Alignment faking in Large Language Models", and is currently working on a third party investigation into the OpenAI/HuggingFace incident. In my opinion, he's one of the most interesting thinkers on the future of AI. Had him on to discuss/debate recursive self-improvement. This might be the most important question in the world right now – whether within a year or so of achieving human-level intelligence, you slingshot towards having 10s of billions of superintelligences, each of which is dramatically more competent than human experts across all fields. I’ve historically been skeptical of this possibility. My intuition has been that we will end up significantly bottlenecked by not only compute scaling but human expert data, which I think underlies most of the AI progress today. If, because of RSI, we got a jump as big as GPT-3 to a Mythos (i.e. 6 years of AI progress) within a single year of achieving AGI, then the thing we get there at the end of that year is definitively and wildly superhuman. We hashed it out, and I think Ryan made a pretty…

People in this episode

Host: Dwarkesh Patel

Guest: Ryan Greenblatt

Topics covered

• AI research
• technical AI safety
• recursive self-improvement
• human-level intelligence
• superintelligences
• alignment implications

Keywords

• AI safety
• recursive self-improvement
• superintelligence
• AGI
• alignment
• Redwood Research
• OpenAI
• HuggingFace
• AI progress

Mentioned in this episode

Organizations: Redwood Research, OpenAI, HuggingFace

More episodes of Dwarkesh Podcast

• Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging Face · September 1, 2026 · 2h 21m
• The rise and fall of agent civilizations · August 31, 2026 · 25 min
• Dylan Patel – Anthropic & OpenAI will have most of the world’s compute by 2028 · August 25, 2026 · 1h 17m
• 8 Predictions for the Era of Continual Learning · August 7, 2026 · 9 min
• Why smarter AI models could drive up compute prices 10x · August 3, 2026 · 11 min

Explore listener stats, chart rankings, contacts and more on the Dwarkesh Podcast podcast page.