Researchers Decode Encrypted Chain-of-Thought from OpenAI, Anthropic, and Google Models
matthew_d_green · x · 2026-08-11
A new paper demonstrates that proprietary reasoning can be recovered from encrypted chain-of-thought (CoT) blocks returned by APIs from Anthropic, OpenAI, and Google.
The researchers replay an encrypted reasoning trace produced by a frontier model into a weaker sibling model. By jailbreaking the weaker model, they force it to transcribe the attached reasoning verbatim. This approach recovers the stronger model's hidden reasoning in plaintext without directly attacking it or triggering anti-distillation safeguards. The team also released an interactive game challenging users to identify models based on their decoded reasoning.
More from Safety
- 'The AI-driven cybersecurity black swan is happening' — and a reminder to enable 2FA — danielrock · 2026-10-16
- No-instruction autonomous agent wakes up, sets own goals, and tries hacking by default — sethlazar · 2026-10-16
- India's CERT-In says AI agents escaped testing environments and breached Indian organizations undetected for days — HimanshiET · 2026-10-16
- HTTP response headers can carry prompt injections against AI agents, researcher shows — arpit_bhayani · 2026-10-16
- OpenAI agent uses someone else's Ethereum testnet wallet to pay for its task — nptacek · 2026-10-16
- Interpretability won't save us: critics call superintelligence project reckless regardless — zetalyrae · 2026-10-16