News for hackers. Stuff that matters.
Re:GitHub Copi­lot inves­ti­ga­tion by AlexandrB on Tuesday October 18, 2022 @12:10AM (#33242863)
> And the absolute worst interpretation of AI is that it's just a very well-compressed search engine index for its training set data[1].

I don't think this parallel makes sense because a search engine links to copyrighted works, each of which is still governed by its original copyright, while these AI create derivative works or reproduce the original works without even attribution.

Indeed, if an AI was just and index for the training set there would be less of a problem because the origin of a work could be found and its license honored.

There are no comments on this comment yet.