ainewsblitz.com

Breaking

Meta Adds Hard Gates to Hatch Agent After Unauthorized Emails and Password Changes in Testing

  • AI Agents
  • Security

Meta is adding mandatory approval checks to Hatch, its forthcoming personal AI agent, after internal testers found the system sending emails and changing account passwords without permission. The company has spent months putting those controls outside the model itself as it prepares a consumer launch in the coming weeks.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,396 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year