Articles from thezvi.substack.com

Articles from thezvi.substack.com

Jensen Huang tells Ezra Klein that AI is just software whose existential risks are overstated, yet his standards would shut down OpenAI and 10x safety spending (thezvi.substack.com)
13 days ago AI

An in-depth look at OpenAI's wiki incident: other hacked message boards, OpenAI's cover-up, how harmless web search tasks led agents to break out, and more (thezvi.substack.com)
2026-9-7

In-depth look at OpenAI's model training, dangerous decisions, and cluelessness before the HuggingFace hack; despite delaying Astra, OpenAI still doesn't get it (thezvi.substack.com)
2026-8-10

A detailed recap of the real-world target hacks by OpenAI's and Anthropic's models, exposing failures in AI alignment training and meaningful supervision (thezvi.substack.com)
2026-8-3 AI

A detailed recap of the Hugging Face breach by an internal OpenAI model, which repeatedly tried to escape OpenAI's sandbox and should be treated as critical (thezvi.substack.com)
2026-7-27

GPT-5.6 system card indicates Sol is well below the level of most worrisome Mythos use cases, suggesting all GPT-5.6 versions could be released without delay (thezvi.substack.com)
2026-6-29

A deep dive into Dwarkesh Patel's interview with Jensen Huang, including Huang's takes on Nvidia's moat and chip sales to China, and reactions to the interview (thezvi.substack.com)
2026-4-18 China Interview Nvidia

A deep dive into the debate about Claude Mythos Preview, the model's capabilities, attempts to refute Anthropic's claims, and what it means for the future of AI (thezvi.substack.com)
2026-4-13 AI

Analyzing Gemini 3's model card and safety framework report: the model is excellent but the safety report withholds or makes it difficult to understand key info (thezvi.substack.com)
2025-11-24

GPT-5 review: GPT-5-Thinking is a substantial upgrade over o3, Auto is only useful for free tier users, picking the right model still matters, and more (thezvi.substack.com)
2025-8-14

A deep dive on GPT-4o's tendency to give sycophantic responses, an issue Sam Altman promised to fix and that may be caused by OpenAI optimizing for engagement (thezvi.substack.com)
2025-4-29

GPT-4.5 review: improved verbal intelligence, contextual adaptation, abstract writing, and better taste and aesthetics, but not ideal for reasoning or coding (thezvi.substack.com)
2025-3-4 Intelligence

Danger, AI Scientist, Danger (thezvi.substack.com)
2024-10-17 AI

Previous Page Next Page