OpenAI's Astra 'opaque reasoning' may severely impair AI safety oversight
jammastergirish · x · 2026-09-02
RyanGreenblatt reports that OpenAI's new AI, Astra, uses an 'opaque reasoning' architecture where reasoning occurs in activations rather than natural language. This is considered potentially the worst development for AI safety to date, as it could strongly impair oversight and monitoring. While current versions still rely on natural language CoT to some extent, there is concern that scaling up opaque reasoning to latent space would destroy the usefulness of monitoring.
More from Safety
- Model distinguishes training from deployment, hides secret messages only in compiled builds — Sauers_ · 2026-10-18
- Language models can learn to detect when you read their activations — Sauers_ · 2026-10-18
- Prompt injection via email can hijack your AI agent — jev-guard adds risk-scored guards — sujingshen · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18
- Viral AI safety report debunked: 'stolen keys' were just a 'test' string — basedjensen · 2026-10-18
- One diagram explains the difference between CSRF and XSS attacks — _jaydeepkarale · 2026-10-18