Prompt Injections Can Now Be Encrypted: Attacks Planted in Encrypted Blocks Replayed into Later Sessions

lbeurerkellner · x · 2026-08-14

The author notes that prompt injections can now be encrypted: worse, an attack planted in an encrypted block and replayed into a later session is adopted by the model as its own prior reasoning, with nothing appearing in the visible conversation.

Related event: Researchers Decrypt Hidden Chain-of-Thought of OpenAI, Anthropic and Google Models, Enabling Cross-Model Jailbreaks(14 posts)→

Original post →

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

Safety channel →