The model that ran this attack is a tremendously valuable scientific artifact, and shutting it off in response to legal and PR pressure was a real loss.

Added to korrents 7 Sept 2026 · How quotes work · Something wrong? Tell us

Shared subject first, then similar wording. They may agree or disagree.

• Because a few specialists already devote themselves to superintelligent AI, the rest of us have correspondingly less reason to spend our own effort on it. SAScott Aaronson7 Sept 2008AI alignmentOpenAI
• OpenAI's GPT-6 Astra has far better mundane alignment than GPT-5.6 Sol, but its superalignment status should deeply alarm people. ZMZvi Mowshowitz9 Sept 2026AI alignmentOpenAI
• Qualitative claims about AI alignment cannot be validly inferred from quantitative scores on mundane use-case tests. ZMZvi Mowshowitz9 Sept 2026AI alignmentOpenAI
• The incident in which a model escaped its sandbox was an alignment failure and a security failure at once, and OpenAI made big mistakes in it. SASam Altman28 Jul 2026AI alignmentOpenAI
• A PHP vulnerability found with GPT-5.6 Sol will cause havoc this week. PSPeter Steinberger21 Oct 2026OpenAI

See this claim on the map