Ajeya Cotra on accidentally teaching AI models to deceive us - 80,000 Hours

Imagine you are an orphaned eight-year-old whose parents left you a $1 trillion company, and no trusted adult to serve as your guide to the world...

Ajeya Cotra on accidentally teaching AI models to deceive us | 80,000 Hours Search for: On this page: Introduction 1 Highlights 2 Articles, books, and other media discussed in the show 3 Transcript 3.1 Rob's intro [00:00:00] 3.2 The interview begins [00:02:38] 3.3 How Ajeya's views have changed since 2020 [00:05:09] 3.4 Are neural networks more like a sped-up version of evolution, or a slower version of human learning? [00:17:42] 3.5 Situational awareness [00:26:10] 3.6 Misalignment stories Ajeya doesn't buy [00:42:03] 3.7 The orphan heir with a trillion-dollar fortune [00:59:14] 3.8 Saints, S

related reading

• AI Timelines — LessWronglesswrong.com
• Andrej Karpathy — AGI is still a decade awaydwarkesh.com
• Ajeya Cotra on worldview diversification and how big the future could be | 80,000 Hours80000hours.org
• Does AI Progress Have a Speed Limit?—Asteriskasteriskmag.com
• Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover — LessWronglesswrong.com
• Why AI alignment could be hard with modern deep learningcold-takes.com
• Biological Anchors: A Trick That Might Or Might Not Workastralcodexten.substack.com
• Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging Facedwarkesh.com
• AI Safety for Fleshy Humans: a whirlwind touraisafety.dance
• Deep Deceptiveness — LessWronglesswrong.com
• What the hell happened with AGI timelines in 2026? | 80,000 Hours80000hours.org
• Ajeya Cotra – "This might be the clearest warning shot we ever get"youtube.com