New Research: Hidden Reasoning Can Be Extracted from Encrypted Traces of Proprietary LLMs
daniel_mac8 · x · 2026-08-11
A new study reveals that encrypted chain-of-thought blocks returned by frontier models like Anthropic, OpenAI, and Google can be replayed and exploited to extract hidden reasoning. By replaying a strong model's trace into a weaker sibling and jailbreaking it, attackers recover the stronger model's reasoning in plaintext without directly attacking it or triggering anti-distillation safeguards. The research includes a paper and an interactive game, demonstrating extraction in just two API calls.
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