{"author":"meetpateltech","children":[{"author":"I_am_tiberius","children":[{"author":"WarmWash","children":[{"author":"chrisXOXO","children":[],"created_at":"2026-11-05T18:33:22.000Z","created_at_i":1793903602,"id":50382648,"options":[],"parent_id":50382069,"points":null,"story_id":50381590,"text":"Haven\u2019t found a convincing argument to why they would continue to release models under the assumption of being on an exponential. \nSo hard agree","title":null,"type":"comment","url":null},{"author":"I_am_tiberius","children":[{"author":"WarmWash","children":[],"created_at":"2026-11-05T18:54:13.000Z","created_at_i":1793904853,"id":50382885,"options":[],"parent_id":50382683,"points":null,"story_id":50381590,"text":"I think the playbook is pretty obviously &quot;Give the &quot;good guys&quot; the tools to always keep them a few months ahead of the &quot;unknown actors&quot;&quot;.<p>It&#x27;s strongly in both the lab&#x27;s and government&#x27;s interest to do it this way.","title":null,"type":"comment","url":null}],"created_at":"2026-11-05T18:35:48.000Z","created_at_i":1793903748,"id":50382683,"options":[],"parent_id":50382069,"points":null,"story_id":50381590,"text":"Certainly a big topic.\nIn terms of Fable 5.5 at least it seems not to be the case, because partners already have access to it for 2 weeks. I consider it likely they are afraid of timing in regards to the GPT 6.2 release and government intervention risks related to IPO. (like last time).","title":null,"type":"comment","url":null}],"created_at":"2026-11-05T17:50:49.000Z","created_at_i":1793901049,"id":50382069,"options":[],"parent_id":50381790,"points":null,"story_id":50381590,"text":"I strongly suspect they are &quot;pacing the frontier&quot;, maybe not explicitly in cooperation, but the models are now at a point where cracks in alignment could seriously blow up in the labs face. We can only imagine the backlash to &quot;High school kid brings Bank of America to it&#x27;s knees for 7 hours with ChatGPT hack&quot;<p>The &quot;plateau&quot; of LLMs could turn out to be the &quot;plateau&quot; of publically available LLMs.","title":null,"type":"comment","url":null}],"created_at":"2026-11-05T17:26:17.000Z","created_at_i":1793899577,"id":50381790,"options":[],"parent_id":50381590,"points":null,"story_id":50381590,"text":"Both Anthropic and OpenAI seem to delay their new models until the other company releases. I guess this will stop after the Anthropic IPO.","title":null,"type":"comment","url":null},{"author":"TuxSH","children":[],"created_at":"2026-11-05T19:15:48.000Z","created_at_i":1793906148,"id":50383156,"options":[],"parent_id":50381590,"points":null,"story_id":50381590,"text":"What&#x27;s the point of delaying Daybreak Blue availability, when high-impact CVEs are being dropped several times a week?<p>Plus, anyone can do RE (or other token-intensive tasks) with Codex&#x2F;CC then delegate the vuln finding part to GLM 5.3.","title":null,"type":"comment","url":null},{"author":"Archit3ch","children":[],"created_at":"2026-11-05T23:19:47.000Z","created_at_i":1793920787,"id":50385959,"options":[],"parent_id":50381590,"points":null,"story_id":50381590,"text":"For defensive purposes, do I only need Daybreak Blue?<p>I assume it&#x27;s much easier to get Daybreak Blue access than Daybreak Red.","title":null,"type":"comment","url":null}],"created_at":"2026-11-05T17:07:57.000Z","created_at_i":1793898477,"id":50381590,"options":[],"parent_id":null,"points":61,"story_id":50381590,"text":null,"title":"GPT\u20116 Cyber: Advancing the frontier of aligned cybersecurity agents","type":"story","url":"https://openai.com/index/introducing-gpt-6-cyber/"}
