Not quite. Before commercial forensics tools started pixelating and obfuscating images so investigators didn't have to witness CSA constantly, they'd get desens

Not quite. Before commercial forensics tools started pixelating and obfuscating images so investigators didn't have to witness CSA constantly, they'd get desensitized to this stuff and occasionally end up spiraling. We're seeing it again with Facebook moderators and AI trainers. I haven't heard any new cases of exfiltration in a very long time (at worst, just vice cops soliciting prostitutes) so I like to think the technical/procedural changes are working as intended.