Zvi Mowshowitz on AI
7 quotes from Zvi Mowshowitz on the wall, 2023 to 2026, newest first. Each one is verbatim and links to where it was said. Sources include thezvi.substack.com.
- 2026
Zvi Mowshowitz rebutting a WSJ headline claiming China matched Anthropic's AI in cybersecurity, June 2026.
thezvi.substack.comWSJ Article Claiming China Has Matched Anthropic Is Obvious Nonsense
Users are increasingly not supervising the work, that's the whole idea
Zvi Mowshowitz responding to OpenAI's warning that users should supervise GPT-5.6 as a coding agent, in his review of the GPT-5.6 system card.
Claude could do the thing, it simply decided not to do the thing
Zvi Mowshowitz, writing about Anthropic's Claude Opus 4.8 system card, describing observed failure modes where the model had the capability but did not perform the task correctly.
It might be a good model, sir. but no
Zvi Mowshowitz's assessment of DeepSeek v4 in his weekly AI roundup, arguing it is not a frontier model despite engineering merits.
- 2025
Potentially the largest theft in human history
Zvi Mowshowitz's characterization of OpenAI's nonprofit-to-PBC restructuring, written on his Substack in October 2025.
thezvi.substack.comOpenAI Moves To Complete Potentially The Largest Theft In Human History
- 2024
Zvi Mowshowitz, in his analysis of DeepSeek v3's technical report, contrasted its detailed engineering approach with Meta's Llama paper.
thezvi.substack.comDeepSeek v3: The Six Million Dollar Model
- 2023
This market does not think about you at all
From Zvi Mowshowitz's blog post 'On AI and Interest Rates', January 2023, discussing why interest rates fail to reflect beliefs about transformative AI.