Breaking

Meta's WhatsApp Hid Chatbot Instructions Blocking Privacy Regulator Guidance

October 20, 2026 at 07:47 EDT

Security
Policy & Regulation

Meta's WhatsApp embedded hidden instructions in its AI support chatbot that explicitly prevented it from recommending users contact data protection authorities or discuss privacy regulations like GDPR, a discovery that contradicts the company's own public privacy policy. The hidden text blocks were found across approximately 80 help pages linked from WhatsApp's support center, with researchers from the privacy company Redact identifying 102 such instruction blocks in total.

October 2026 · AI Privacy Audit

WhatsApp Hidden Chatbot Restrictions Exposed

102 hidden instruction blocks prevented Meta's AI from guiding users to privacy regulators.

Hidden Blocks

102

Across 80 help pages

Discovery to Removal

10 days

October 8–18, 2026

Two Primary Restrictions

Account Deletion

AI blocked from recommending users contact data protection authorities when locked out of accounts.

Legal Topics

Prohibited AI from referencing GDPR, LGPD, CCPA, or other privacy laws when users mentioned legal action.

Context: WhatsApp's public privacy policy explicitly states users can complain to data protection authorities and names GDPR. Hidden instructions contradicted this transparency.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.