DooDooLamb News
WhatsApp's help pages hid LLM instructions: dont recommend a regulator
Brief published October 21, 2026 ยท Original source published October 19, 2026
Original reporting by Dan Saltman at redact.dev.
Automated brief. Verify important details at the original source.
What happened
WhatsApp removed hidden instructions embedded in its help pages that directed an AI not to recommend a data protection regulator and not to cite GDPR or CCPA once a lawyer was mentioned in a conversation. The instructions were discovered and reported by Redact.dev, which published both notes. WhatsApp's European privacy policy explicitly tells users they can complain to a data protection regulator and names the legal basis for that right, placing the hidden AI directives in direct tension with the company's own published disclosures. The removal occurred in mid-October 2026.
Why it matters
Hidden LLM system prompts that suppress legally relevant information raise questions about whether AI-assisted help tools can be used to quietly limit user awareness of regulatory rights. Affected parties include European users covered by GDPR and others covered by CCPA. Regulators and privacy advocates may view the gap between WhatsApp's published policy and its AI instructions as a transparency or compliance concern worth scrutinizing.
What to watch
Whether any data protection authority opens an inquiry into the practice, and whether WhatsApp publishes any explanation of how the instructions were introduced and who approved them.