Meta Accidentally Leaks WhatsApp AI Rules, Including What It Should ‘Not Say’ To Users

Meta Accidentally Leaks WhatsApp AI Rules, Including What It Should ‘Not Say’ To Users

Meta might have revealed some of the internal information that it was not supposed to. WhatsApp accidentally exposed internal instructions meant for one of Meta’s AI-powered support chatbots on dozens of its help and FAQ pages. This hidden text apparently included information about how the chatbot was told to react to sensitive questions. It also involves specific directions on what it should not tell users about legal action, privacy regulations and upcoming WhatsApp features.

‘What Not To Say’: WhatsApp Accidentally Exposes Hidden Instructions For Meta AI Chatbot (AI-generated)

As per the analysis by privacy company Redact, first reported by Wired, around 80 pages on WhatsApp’s website contained instructions for a large language model (LLM). These directions were not displayed as part of the normal help articles but could be found in the underlying code of the webpages.

Redact reportedly discovered nearly 100 blocks of hidden text across WhatsApp’s help pages. Some of the instructions were labelled “WHAT NOT TO SAY” and appeared to tell the AI chatbot how to handle particular questions or situations. The instructions appeared in code blocks carrying the header “HelpContentGalaxyLLMInclude.”

Dan Saltman, CEO and founder of Redact.dev, told Wired that many of these instructions were relatively straightforward and directed users towards official support options. However, the material did not seem to have been intended for public viewing.

“The stand-out thing here is the gap between what people are told vs LLMs,” Saltman said. Some of the most detailed instructions concerned users who raised legal issues while seeking WhatsApp support.

On WhatsApp’s “About reporting and blocking on WhatsApp” page, the instructions reportedly included specific directions for situations where a “user mentions legal action, lawsuits, lawyers, attorneys, courts, or litigation.”

After Meta was contacted about the exposed material, the instructions were removed from the affected WhatsApp pages. The “HelpContentGalaxyLLMInclude” text is no longer visible in the code of the live pages.

Govind Choudhary is the Chief Copy Editor for Tech at Times Now with over … Govind Choudhary is the Chief Copy Editor for Tech at Times Now with over five years of experience in the media industry. He covers consumer technology, gadgets, and emerging digital trends. A gadget enthusiast and avid content consumer, Govind is also passionate about bikes and enjoys exploring different cuisines in his free time. View MoreView Less

📰 Original Source Attribution

Reported by timesnownews.com.

Read Original Report at timesnownews.com ↗
Share: WhatsApp WhatsApp
💬

Comments (0)

Join the Conversation

No comments yet. Be the first to share your opinion!

You may like