(twitter.com)

229 points modinfo | 1 comments | 30 Jun 24 01:56 UTC | HN request time: 0.211s | source

Show context

Klonoar ◴[30 Jun 24 08:17 UTC] No.40835818[source]▶

Alright, here’s a take I haven’t seen in this thread yet: how could this be used for fingerprinting, beyond an existence check for the API itself?

replies(2): >>40835874 #>>40836402 #

INTPenis ◴[30 Jun 24 08:32 UTC] No.40835874[source]▶

>>40835818 #

This assumes the model is different on each computer.

And that made me realize that Google might start training it with your browser history. Anything is possible at this point.

replies(1): >>40836334 #

1. TonyTrapp ◴[30 Jun 24 10:30 UTC] No.40836334[source]▶

>>40835874 #

Even if the model data is the same - we have seen that fingerprinting can be applied to WebGL, so if hardware acceleration is used to run those models, it might be possible to fingerprint the hardware based on the outputs?

↑

Chrome is adding `window.ai` – a Gemini Nano AI model right inside the browser