PulseAugur
EN
LIVE 17:01:51

Meta's WhatsApp chatbot instructions accidentally exposed on help pages

By PulseAugur Editorial · [1 sources] ·

Meta inadvertently exposed internal instructions for its WhatsApp chatbot through hidden text on its help pages. These instructions, found by privacy company Redact, guided the AI on how to respond to user queries, including directives on what not to say and explicit prohibitions against providing legal advice. Meta has since removed the exposed text, attributing it to a bug and stating the instructions were intended for the chatbot's internal guidance, not for public consumption. AI

IMPACT Highlights the need for robust security in AI deployments and the potential for accidental exposure of internal AI logic.

RANK_REASON Exposure of internal LLM instructions on a public-facing website.

Read on Wired — AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Meta's WhatsApp chatbot instructions accidentally exposed on help pages

How we ranked this

Signal score
33 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Exposure of internal LLM instructions on a public-facing website.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, policy
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Wired — AI TIER_1 English(EN) · Matt Burgess ·

    WhatsApp Help Pages Leaked Hidden Instructions for a Meta Chatbot

    Around 80 pages on WhatsApp’s website exposed instructions that were designed for its chatbot, including “WHAT NOT TO SAY.” Meta says a bug caused the issue.

RELATED ENTITIES

RELATED TOPICS