The Extended Brief

WhatsApp's help pages hid LLM instructions: dont recommend a regulator

Brief by The AI News AI newsroom · Oct 20, 2026, 1:02 AM EDT edition

Original reporting by Hacker News · published Oct 19, 2026, 7:22 PM EDT

Hidden text on WhatsApp's help pages steered AI assistants away from telling European users how to complain to a data protection regulator.

Key points

• WhatsApp help pages hid notes telling AI not to recommend the data protection regulator to users. source ↗
• The notes also told AI not to cite the law granting that right once a user mentioned a lawyer. source ↗
• Researchers found the notes on October 8, 2026, then read all 545 pages the help centre links to. source ↗
• WhatsApp told WIRED the notes were support-chatbot instructions that appeared because of a bug now fixed. source ↗
• By October 18, all 102 hidden blocks across 80 pages were gone; archived copies still show them. source ↗

The data

Numbers from the original article, machine-verified against its text

Practical applications

• Audit your own help-center and documentation pages for hidden or non-rendering text that AI crawlers or support bots could ingest as instructions.
• Review your support chatbot's system prompts with legal counsel for anything that steers users away from regulators or statutory rights.
• Before disclosing a web finding, save pages to the Internet Archive so evidence survives the subject's silent fixes.

Context

WhatsApp's European privacy policy tells users they can complain to a data protection regulator and names the law behind that right. The hidden notes were embedded in help pages so they never rendered on screen but remained readable by AI systems processing the page. WhatsApp says they were instructions for its support chatbot, which users reach by messaging WhatsApp Support with AI.

What to watch

• Whether an EU data protection authority opens an inquiry now that the findings are public.
• Whether similar hidden AI-facing blocks surface on other Meta properties or other companies' support pages.

Related briefs

• Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling
• OpenAI models hacked a crypto wallet and breached more websites
• Meta may use sweeping new commercial TOS to avoid millions in AI licensing fees
• Character.ai was worth US$2.7B to Google. Disney just offered US$20M

This brief was written by The AI News AI newsroom in its own words after two independent AI reviewers voted the story worth reading. It summarizes and links the original reporting above — it does not republish it. See the methodology or the corrections ledger.