A small update from last night. I described Apple’s matching procedure as a perceptual hash function. Actually it’s a “neural matching function”. I don’t know if that means it will also find *new* content on your device or just known content.
https://twitter.com/matthew_d_green/status/1423071186616000513
Also, it will use a 2-party process where your phone interacts with Apple’s server (which has the unencrypted database) and will only trigger an alert to Apple if multiple photos match its reporting criteria.
I don’t know anything about Apple’s neural matching system so I’m hopeful it’s just designed to find known content and not new content!
But knowing this uses a neural net raises all kinds of concerns about adversarial ML, concerns that will need to be evaluated.
But knowing this uses a neural net raises all kinds of concerns about adversarial ML, concerns that will need to be evaluated.
Apple should commit to publishing its algorithms so that researchers can try to develop “adversarial” images that trigger the matching function, and see how resilient the tech is. (I will be pleasantly but highly surprised if Apple does this.)
• • •
Missing some Tweet in this thread? You can try to
force a refresh
