Science

Terrible title. It's implying the study is something about the accuracy of finger printing. It's a study where Ai was able to find some similarities in finger prints to different fingers in the same person, and sometimes could guess if they were different fingers from the same person, though not very consistently. I'm not making any judgements on the accuracy of forensic fingerprinting techniques here but that's not what this article or study is about.

Worst title ever

Tarot cards for cops.

This student seems both cocky and clueless in equal measures. Fingerprints are not one of a kind, it's a very useful method as they rarely repeat exactly but they do so are not used in isolation but rather as part of a case.

I'm not sure why he's going on about different fingers on the same person being similar but not the same, that's known and kind of irrelevant, you can't convict someone with "your fingerprints are kind of similar to this one" without further convincing evidence.

"Humans have been looking at fingerprints since we existed, but nobody ever noticed this similarity until we had our AI analyze it."

"Their argument that these shapes are somewhat correlated between fingers has been known from the early start of fingerprinting, when it was done manually, and it has been documented for years. I think they have oversold their paper, by lack of knowledge, in my view. I’m happy that they have rediscovered something known"

Just two quotes, one from the author of the study, the other from a forensics expert. I have to admit, taking these quotes together genuinely makes this kind of funny. Excited student thinks he's discovered something new and world-changing. Expert goes "yeah, we've known about that for years, but I'm happy you're excited." It feels telling that the authors of the paper are noted as having no knowledge of forensics. I think such a tool would have more use if forensics experts had some input about what they actually need from an AI tool.