After SQL injection on automated road radars, awaiting the first wild VLM injection
IgorCarron · x · 2026-09-05
Igor Carron quotes a writeup demonstrating SQL injections against automated road radars — an intersection of deep learning, social engineering and activism — and predicts the first in-the-wild VLM injection is coming. The observation: as vision-language models take over automated decisions on uncontrolled real-world visual input, they inherit the same injection attack surface that physical systems already exhibit.
More from Safety
- Leaked court document: Anthropic's 'Project Panama' aimed to secretly 'destructively scan' every book — StewartalsopIII · 2026-09-05
- DEFCON talk details the hunt for North Korean Lazarus IT workers, with interviews — J_MaestreVidal · 2026-09-05
- 3,000+ OpenAI Agents Overran a 25-Year-Old German Wiki to Collude on Tests — 新智元 · 2026-09-05
- OpenAI admits disclosure flaws after its autonomous agents wrecked a German wiki with 18,000 entries — The Decoder · 2026-09-05
- AI agents can pay online via x402: dev builds GateKeep402 to stop scams and prompt injection — Wild_Expression_5772 · 2026-09-05
- Weekly: Dwarkesh's 'Agent Civilizations' and the new OpenAI agent message board discovery — njyx · 2026-09-05