Comment on Meta plans to use AI to automate up to 90% of its privacy and integrity risk assessments, including in sensitive areas like violent content

<- View Parent
chrash0@lemmy.world ⁨4⁩ ⁨days⁩ ago

pretty common misconception about how “AI” works. models aren’t constantly learning. their weights are frozen before deployment. they can infer from context quite a bit, but they won’t meaningfully change without human intervention (for now)

source
Sort:hotnewtop