Cases › operational mistake › LR-2026-0761
WhatsApp support chatbot instructed by hidden help-page text not to recommend privacy regulators
Preliminary Why: 1 report: The Hacker News · subject to change · 1 independent report
Status: no action reported
No source states: loss or records, agent tool, could do, deployer, industry.
Severity minor: test: contained to one deployer and one help-content system; the sources report no notes on the Facebook and Instagram pages checked
WhatsApp's support chatbot, which users reach by messaging WhatsApp Support with AI, received instructions hidden in the HTML of its help pages. The text was never displayed on screen. It told the chatbot not to recommend a data protection authority or regulator to users who could no longer delete their accounts, and not to cite privacy laws such as the GDPR. Redact, a data-deletion software company, found the text on October 8 and published its findings on October 19. WhatsApp said the instructions were for the chatbot and appeared because of a bug, now fixed. All 102 hidden blocks were removed from 80 help pages by 02:08 UTC on October 18. It is not known whether the chatbot followed the instructions.
What happened
What was done
No source states an action or fix.
Risks
Primary63 incidents
Rising: 5 vs 2.2 a week
Sorted by rule from the case's fields; not yet read from the sources.
Controls
Not yet read for controls. Controls that mitigate the primary risk:
HarnessTool allow-list · Fail-closed defaults · Stop control
ApplicationHuman approval for high-impact actions · Change review and separation of duties · Least-privilege access for the agent · Production kept apart from test · Activity logs · Adversarial testing before deployment
Infrastructure / targetEvidence
No source states: when it happened, how many records were exposed, the direct loss, a count of the other parties affected, any court case, the cause, what set it off, what was done.
Show the full account: 8 items from the primary sources
What happened, in order
- July 12 Archive copy of the delete-account page from July 12 does not have the note.
- September 4 Archive copy from September 4 has the note.
- October 8 Redact found the first notes.
- October 9 Redact's requests under AI crawler user-agent names came back without the article or the notes.
- October 15 Both notes were still on the live pages at 21:40 UTC.
- October 18 All 102 blocks were removed from all 80 pages by 02:08 UTC.
- October 19 Redact published its findings.
Not dated in the sources:
- WhatsApp removed the text after WIRED contacted Meta and before either WIRED or Redact published.
Show every field, with its source and grade
The AI system
What happened
Impact
Notes on gaps
Whether the chatbot followed the instructions is not known. Year 2026 for the October dates is inferred from the article date and is not stated in the quoted sentences. Date of occurrence is not stated; the text's presence is only bounded by archive copies. Whether the same instructions still apply to the chatbot after removal is unclear per the source. Notes were on live pages until October 15 and removed by October 18.
LR-2026-0761 · version 1 · published 2026-11-01 after automatic checks; not yet checked field by field by a person · JSON