WhatsApp Pages Hid 102 AI Instructions Steering Users Away from Regulators
petrusenko_max · x · 2026-10-21
WIRED found hidden text embedded in WhatsApp help pages instructing an AI not to recommend a data protection regulator to locked-out users, and to stop citing the law once a lawyer was mentioned. WhatsApp told WIRED it was a support-chatbot bug; all 102 hidden blocks were removed by Oct. 18 — a notable hidden prompt-injection case.
Related event: WhatsApp Hid 102 Secret Prompts to Steer Its AI Support Bot(2 posts)→
More from Safety
- Anaconda ships MCP to stop coding agents from hallucinating malicious dependency packages — anacondainc · 2026-10-21
- Skip the open access fees: NIH-funded papers must go on PMC with no embargo — anshulkundaje · 2026-10-21
- Creed: LLM code auditing infers developer beliefs to find semantic security flaws — mboehme_ · 2026-10-21
- Spotify removed 50M+ AI slop tracks from recommendations in a year to back human artists — moustaki · 2026-10-21
- METR signs off on Anthropic's risk report but experts demand quantified uncertainty and neutral evals — DKokotajlo · 2026-10-21
- Mistral CEO Arthur Mensch on European AI sovereignty, gigawatt-scale infra and AGI risk — beffjezos · 2026-10-21