Meta's WhatsApp AI chatbot instructions accidentally exposed on help pages
Meta accidentally exposed internal instructions for its WhatsApp AI chatbot on public help pages, which were subsequently removed. The leaked guidelines detailed what the chatbot should and should not say, including directives to avoid legal, medical, or financial advice. This incident was widely reported by multiple news outlets after the discovery of the hidden content.
4 articles from 4 outlets covered this story. The underlying claim is sourced from a shipped.
What do all outlets agree on?
4 outlets covered “Meta's WhatsApp AI chatbot instructions accidentally exposed on help pages”. All of them report the following:
- Meta's WhatsApp AI chatbot instructions were exposed
- Instructions found on WhatsApp help pages
- Exposure was accidental
- Guidelines detailed what chatbot should and should not say
- Instructions included avoiding legal, medical, or financial advice
- Help pages were subsequently removed
Which outlets covered this?
All 4 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.
safety1
ethics3