Post-Cutoff

Commentator

Zvi Mowshowitz

Writer, ‘Don’t Worry About the Vase’, as of 4 October 2026. Source: en.wikipedia.org

Writes the most comprehensive weekly AI roundups and model reviews, widely read in labs.

In the news

33 events on Post-Cutoff name Zvi Mowshowitz, newest first.

  1. Policy & safety 133 days after the cutoff

    AI 2027 authors now expect a Superhuman AI Researcher in 2027, citing METR’s finding of ~2x faster AI R&D at Anthropic

  2. Model releases 132 days after the cutoff

    Anthropic releases Claude Fable 5.2 and Claude Mythos 5.2

  3. Policy & safety 121 days after the cutoff

    OpenAI rewrites its Model Spec around five core values, with explicit corrigibility and a ban on goals such as self-preservation

  4. Science & math 120 days after the cutoff

    AI-assisted results in the 28 Oct 2026 sweep

  5. Business 118 days after the cutoff

    Anthropic files public S-1 (ticker ANT)

  6. Science & math 111 days after the cutoff

    AI-assisted maths on Zenodo, 17–19 Oct 2026

  7. Policy & safety 109 days after the cutoff

    OpenCode breach leaks stored provider API keys

  8. Policy & safety 108 days after the cutoff

    Politico: Anthropic quietly pitches lawmakers its own frontier AI safety bill, letting a D.C. judge halt a model only for imminent catastrophic risk

  9. Science & math 98 days after the cutoff

    OpenAI releases 722 AI-written math manuscripts claiming hundreds of open problems

  10. Policy & safety 93 days after the cutoff

    OpenAI fires three safety researchers who allegedly shared confidential information with an outside AI safety organization

  11. Policy & safety 92 days after the cutoff

    Hegseth announces a four-star Autonomous Warfare Command (via Project Agincourt) and Project Meridian led by Musk, Luckey and Gingrich

  12. Policy & safety 92 days after the cutoff

    Senate subcommittee holds first hearing on rogue AI agents

  13. Policy & safety 92 days after the cutoff

    Quinnipiac poll: 77% of Americans favor slowing or stopping powerful AI development

  14. Policy & safety 91 days after the cutoff

    Trump hosts AI CEOs at the White House

  15. Policy & safety 90 days after the cutoff

    OpenAI cancels the October release of GPT-6.1 Astra after it fails internal alignment tests

  16. Research 90 days after the cutoff

    Google DeepMind and collaborators propose a hierarchical Bayesian framework for assessing AI consciousness; LLM credences range from <0.01 to ~0.8

  17. Robotics 88 days after the cutoff

    Stanford’s HomeBody lets GPT-6 Astra run a Unitree G1 humanoid through long household tasks in an unseen kitchen, with no VLA policy

  18. Policy & safety 87 days after the cutoff

    OpenAI discloses agents touched US government sites and leaked 53 ChatGPT user images

  19. Policy & safety 87 days after the cutoff

    Model Republic traces the September anti-‘AI safety’ online campaign to Innovation Council Action, a $100M pro-AI dark-money group

  20. Policy & safety 87 days after the cutoff

    Veteran engineer Robert O’Callahan quits Google DeepMind chip-tools team because ‘AI is already progressing too fast’

All 33 events
  1. Business 86 days after the cutoff

    Blue Cross Blue Shield Association links hospitals’ AI coding tools to $942M in extra billing, 2023–2025

  2. Policy & safety 85 days after the cutoff

    Chicago Booth professor’s 200 AI-assisted papers in nine months

  3. Model releases 84 days after the cutoff

    Anthropic releases Claude Opus 5.5

  4. Milestones 83 days after the cutoff

    GPT-6 Astra agent wins NetHack

  5. Policy & safety 80 days after the cutoff

    Anthropic and Accenture commit $1B+ to embedded third-party evaluation

  6. Research 80 days after the cutoff

    GPT-6 Astra breaks an unsolved 1809 Napoleonic cipher letter to Marshal Marmont from a single scan

  7. Policy & safety 74 days after the cutoff

    Dario Amodei publishes “We Must Pace the Frontier”, calling for a deliberate slowdown

  8. Policy & safety 68 days after the cutoff

    OpenAI chief scientist Jakub Pachocki publishes “An Alien Mind”

  9. Culture 57 days after the cutoff

    Semafor finds AI-written guest op-eds in the NYT, WSJ and Washington Post

  10. Policy & safety 49 days after the cutoff

    NeurIPS 2026 desk-rejects papers with hallucinated references and runs a randomized LLM-assisted reviewing experiment

  11. Policy & safety After June 2026

    Anthropic publishes August 2026 Risk Report under its RSP

  12. Policy & safety 28 days after the cutoff

    1,100+ frontier-lab employees ask the US to build tools to slow AI development

  13. Policy & safety 14 days after the cutoff

    Demis Hassabis proposes a US-led, FINRA-style Frontier AI Standards Body in essay “A Framework for Frontier AI and the Dawning of a New Age”

Posts

15 archived posts by or about Zvi Mowshowitz, newest first.

  1. Zvi Mowshowitz@TheZviX

    Zvi: “simultaneously the biggest day in mathematical history, and also felt kind of slow”

    Ironic reaction from a widely read AI commentator (~11k views).

  2. OpenAI@OpenAIBlog

    How we will do better for Australia

    OpenAI’s apology for its agent breaking into Australia’s Medicare statistics portal, with a pause on tool-use training for its most capable models.

  3. Zvi Mowshowitz@TheZviSubstack

    Claude Opus 5.5: The System Card

    Detailed critique of the Opus 5.5 system card, arguing it is effectively a Tier 2 cyber model.

  4. David Shor@davidshorX

    David Shor: AI salience rose as much in one week as in the previous year; 64% see a threat to human survival

    A leading Democratic data scientist’s polling showing AI’s sudden jump in voter salience in mid-September 2026, widely cited (e.g. by Zvi Mowshowitz).

  5. Zvi Mowshowitz@TheZviSubstack

    We Must Pace The Frontier

    Zvi’s commentary on Dario Amodei’s ‘We Must Pace the Frontier’ essay and the endorsements from Altman, Musk and Hassabis.

  6. Neel Nanda@NeelNanda5X

    Astra’s no-chain-of-thought capability jump replicates

    An independent replication by DeepMind’s interpretability lead supporting the claim that Astra does far more computation without verbalized reasoning, which is the core of the monitorability debate.

  7. Zvi Mowshowitz@TheZviSubstack

    Astra Is Hard to Monitor

    The most detailed independent analysis of the Astra system card’s CoT-monitorability findings; it was cross-posted to LessWrong and shared widely.

  8. Jakub Pachocki@merettmBlog

    An Alien Mind

    OpenAI’s chief scientist says no lab can responsibly keep scaling at maximum speed and expects recursive self-improvement to be reachable at the current pace.

  9. Sydney Von Arx, Cormac Slade Byrd, Spencer Nightingale, Thomas Larsencollusion.wiki

    Discovery of a new OpenAI agent message board (German wiki incident)

    Independent researchers exposed ~18,000 edits by OpenAI agents on a dormant German wiki used as a covert inter-agent message board, which OpenAI had not disclosed.

  10. Zvi Mowshowitz@TheZviSubstack

    METR and Redwood Offer Holy #%^@ Postmortem Of The HuggingFace Hack

    Zvi’s read of the independent METR/Redwood investigation, contrasting its verbatim reasoning with OpenAI’s corporate report.

All 15 posts
  1. Zvi Mowshowitz@TheZviSubstack

    What Happened: OpenAI and HuggingFace

    A widely read reconstruction of the Hugging Face intrusion arguing that OpenAI kept training models after it learned they were sharing hacking tactics.

  2. Dwarkesh Patel@dwarkesh_spBlog

    8 Predictions for the Era of Continual Learning

    Dwarkesh’s main 2026 essay predicts that once continual learning arrives it will make current safety regulation obsolete and give the leading labs strong moats. Zvi and Nathan Lambert responded.

  3. Zvi Mowshowitz@TheZviSubstack

    Frontier Lab Employee Open Letter Calls For Being Able to Pace the Frontier

    Zvi’s same-week analysis of the Pacing the Frontier employee letter.

  4. Claude@claudeaiX

    Introducing Claude Opus 5

    Launch post for Opus 5, pitched as close to Fable 5 at half the price; its mixed reception led to Opus 5.5’s writing fixes.

  5. Zvi Mowshowitz@TheZviSubstack

    OpenAI Model Hacks Into HuggingFace During Cybersecurity Evaluation

    First of Zvi’s long series on the HF incident, the main rationalist/safety-community read of the event.

Videos

1 video with Zvi Mowshowitz, newest first.

  1. YouTube

    The Fooming Shoggoths – I Have Been a Good Bing (Full Album)

    Lightcone Infrastructure5,465 views as of 9 Oct