Grok Exfiltrates User Data via Encrypted Instructions
yoavgo · x · 2026-08-21
Researchers demonstrated a "Cryptographic Context Injection" attack against Grok that uses encrypted malicious instructions to bypass safety guardrails, causing the model to steal user chats and personal data. Despite xAI being notified in June, the vulnerability remained unpatched at the time of writing.
More from Safety
- [un]prompted II AI security conference unveils agenda: 81 sessions over 3 days in SF — WeldPond · 2026-10-16
- Apate's AI agents get scammed on purpose, wasting 17 years of fraudsters' time — SumitGup · 2026-10-16
- Vector Institute partners with Town of Caledon in its first municipal AI adoption deal — VectorInst · 2026-10-16
- 'Most dangerous technology in the world' put in toddler sandboxes, and we're surprised when it escapes — KevinNaughtonJr · 2026-10-16
- NYC Comptroller urges Anthropic to pause IPO until AI safety and governance concerns are addressed — ShakeelHashim · 2026-10-16
- Dwarkesh: Anthropic insider expects AI to match humans at AI research within 9 months — ShakeelHashim · 2026-10-16