Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Show HN: Corral kill every command your agent starts (github.com/cardinal44)
    1 pointsby 1 minute ago
    —discuss
  2. OpenAI Says It Will Not Release Newest A.I. Model Over Safety Concerns (nytimes.com)
    4 pointsby 1 minute ago
    —discuss
  3. Jylus – give AI systems evidence from changing data, not just retrieved chunks (jylus.ai)
    1 pointsby 2 minutes ago
    —discuss
  4. Knockoff: A browser extension that filters pseudo-brand junk out of Amazon (github.com/shpigford)
    1 pointsby 3 minutes ago
    —discuss
  5. Show HN: Jauvex 1.2, two-way voice chat harness for Claude+Codex+Grok+Jev (github.com/reindent)
    1 pointsby 4 minutes ago
    1comments
  6. AI Companies Are Not (Necessarily) Liable for Unintended AI Cyberattacks (sarahconstantin.substack.com)
    1 pointsby 6 minutes ago
    —discuss
  7. What is the best LLM for turbo charging code?
    1 pointsby 7 minutes ago
    1comments
  8. OpenAI Halts Model Release Amid Safety Escalation (techbuzz.ai)
    1 pointsby 10 minutes ago
    —discuss
  9. Its not just the f*cking sandbox (twitter.com/joedaroo)
    3 pointsby 17 minutes ago
    —discuss
  10. AI and the Revenge of the Non-Techies (maroun-baydoun.com)
    1 pointsby 17 minutes ago
    —discuss
  11. AI's next training data: your dodgy gaming skills (wired.com)
    1 pointsby 20 minutes ago
    —discuss
  12. Humanos – Help Building the Human Operating System (tryhumanos.com)
    3 pointsby 23 minutes ago
    —discuss
  13. Making Things to Make Them (aimode.substack.com)
    1 pointsby 25 minutes ago
    —discuss
  14. GPT-6 Astra Is the Best Vision Model We Have Tested (roboflow.com)
    2 pointsby 25 minutes ago
    —discuss
  15. Why embeddings don't solve RAG [video] (youtube.com)
    2 pointsby 27 minutes ago
    —discuss
  16. LLM makes decisions to raise its training scores, and ignore user directives (joinhandshake.com)
    2 pointsby 28 minutes ago
    1comments
  17. AMD Acquires Fei-Fei Li's World Labs for $8.2B (bloomberg.com)
    2 pointsby 28 minutes ago
    —discuss
  18. 1996 chat room simulator connected to Win95 and System 7 web desktops (lolchat.rip)
    4 pointsby 29 minutes ago
    2comments
  19. Prenatal exposure to the plasticizer DEHP increases autism and ADHD (elsevier.com)
    2 pointsby 29 minutes ago
    2comments
  20. Atrocious AI-Written Tests (gruhn.me)
    2 pointsby 30 minutes ago
    —discuss
  21. I thought I was building a C replacement. I was wrong (c3-lang.org)
    6 pointsby 30 minutes ago
    —discuss
  22. Tech Utopians Want to Build a City for the Post-A.I. World (nytimes.com)
    1 pointsby 30 minutes ago
    —discuss
  23. Show HN
    2 pointsby 37 minutes ago
    1comments
  24. China broadens travel curbs to encompass family of top AI talent (business-standard.com)
    2 pointsby 49 minutes ago
    —discuss
  25. Show HN: LightSpeed – Interactive visualizer of time dilation from 0 to C (lightspeed.webland.pl)
    1 pointsby 49 minutes ago
    —discuss
  26. Reverse Engineering How Meta's Muse Shops (caeliai.com)
    1 pointsby 52 minutes ago
    —discuss
  27. Anthropic's IPO prospectus shows AI vision, surging costs (reuters.com)
    39 pointsby 55 minutes ago
    30comments
  28. Nvidia announces AI safety platform (theverge.com)
    2 pointsby 58 minutes ago
    1comments
  29. Holo4: Powering generalist computer-use agents (huggingface.co)
    2 pointsby 58 minutes ago
    —discuss
  30. Pklm-sandbox – A lightweight open-source LogitsProcessor for local LLMs (github.com/theimmortalpython)
    1 pointsby 59 minutes ago
    —discuss

OpenAI Halts Model Release Amid Safety Escalation

1 pointsby 11m agotechbuzz.ai
Open original View on HN
0 comments
A quiet thread, for now.Start the conversation on HN ↗