AI safety fight turns into a censorship debate over who gets to control information

djcows · x · 2026-07-23

An AI safety debate turns into a censorship argument

The exchange argues over whether AI safety measures are genuinely protective or just censorship in disguise. One side says safety measures feel like taking defenses away from users and giving attackers an edge.

The reply pushes back sharply, saying AI safety is being used as a cover for censorship and pointing out the asymmetry: model makers copied training material, then selectively decide what information to give back.

It’s a compact but classic AI-governance argument about security, power and information control.

Related event: AI safety fight turns into a censorship debate over who gets to control information(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →