T4K3.news

Meta fixes serious AI privacy bug

Meta resolved a security flaw that allowed users to view others' prompts and responses.

July 15, 2025 at 08:00 PM

Meta fixes bug that could leak users' AI prompts and generated content

The tech giant fixed the security flaw, netting a security researcher $10,000 for privately disclosing the bug.

Meta resolves serious security bug affecting AI user privacy

Meta has addressed a security issue that allowed users of its AI chatbot to access others' private prompts and generated responses. Sandeep Hodkasia, founder of AppSecure, reported the bug on December 26, 2024, and received a $10,000 bounty for his disclosure. The fix was implemented on January 24, 2025, with Meta confirming no evidence of exploitation was found. Hodkasia revealed that by manipulating unique identifiers assigned to prompts, users could access content not meant for them. This vulnerability highlights concerns around privacy in the rapidly developing landscape of AI technology, particularly as competitors like ChatGPT also face scrutiny for user privacy.

Key Takeaways

✔️

Meta addressed a major privacy vulnerability in its AI chatbot.

✔️

A security researcher received $10,000 for reporting the bug.

✔️

The incident raises questions about data privacy in the AI sector.

✔️

Competition in AI is intensifying, challenging firms to prioritize security.

"The prompt numbers generated by Meta’s servers were easily guessable."

Hodkasia explained the vulnerability's technical aspect affecting security.

"Meta found no evidence of abuse and rewarded the researcher."

A Meta spokesperson confirmed the company's handling of the security issue.

This incident is indicative of broader trends in AI and security. As companies like Meta rush to innovate, they must also prioritize user privacy and security. The ease with which security flaws were discovered underscores the importance of rigorous testing in software development. Users are becoming increasingly aware of these risks and may lose trust in platforms if their data is not adequately protected. As Meta competes with other AI apps in a crowded market, maintaining user confidence is essential for future growth.