News for hackers. Stuff that matters.
Re:New AI tool detects child sexual abuse material with ‘99%... by alexhaber on Sunday August 02, 2020 @09:14PM (#24033216)
Is this claiming to detect 99% of CSAM or claiming to distinguish between CSAM and non-CSAM with 99% accuracy?

Relevant: https://bohemian.ai/blog/99-accuracy-99-lie/

Re:New AI tool detects child sexual abuse material with ‘99%... by jdm2212 on Sunday August 02, 2020 @10:45PM (#24033737)
> Thorn, the non-profit behind Safer, says it spots the content with greater than 99% precision.

Sounds like it's claiming 99% precision, so 99% of positives are true positives.

Recall probably isn't great, because (a) if it were great they'd brag about it and (b) you don't get to 99% precision without sacrificing some recall.

Seems reasonable to aim high on precision, though, to avoid burying NCMEC in mis-classified data and also to avoid wrongly banning users' content.

Re:New AI tool detects child sexual abuse material with ‘99%... by suizi on Monday August 03, 2020 @01:38AM (#24034520)
Re:New AI tool detects child sexual abuse material with ‘99%... by jdm2212 on Monday August 03, 2020 @02:10PM (#24041228)
What in that article specifically makes you not trust Thorn? They seem to me like they're trying to do good work -- preventing child exploitation -- that necessarily involves partnering with organizations whose values won't always exactly line up with theirs.