🕒 2 days ago

Meta AI Judges Privacy Risks, Humans Relegated to Algorithmic Bystanders

Thumbnail for Meta AI Judges Privacy Risks, Humans Relegated to Algorithmic Bystanders
Image & Article: npr

Meta AI now automates nearly 90% of privacy and risk reviews at Facebook HQ, rapidly approving updates for Instagram and WhatsApp since May 2025. Human scrutiny is reserved for only the rarest, most complex cases. One ex-executive warns: 'higher risks' and faster launches may spell trouble.

Product teams answer AI-driven questionnaires, receiving instant decisions on privacy, youth, and integrity risks. The result: engineers, not privacy experts, steer critical updates. One former insider quips, 'engineers are not privacy experts.' Expect some surreal algorithmic whiplash.

Meta’s AI can greenlight changes to violent content and misinformation before humans ever see them.

Source

Still Feeling Curious?

You Might Like

Weirdfeed Picks