OpenAI and Google reportedly used transcriptions of YouTube videos to train their AI models

OpenAI and Google skilled their AI fashions on textual content transcribed from YouTube movies, probably violating creators’ copyrights, in line with . The report, which describes the lengths OpenAI, Google and Meta have gone to to be able to maximize the quantity of information they’ll feed to their AIs, cites quite a few folks with data of the businesses’ practices. It comes simply days after YouTube CEO Neal Mohan mentioned in an interview with that OpenAI’s alleged use of YouTube movies to coach its new text-to-video generator, Sora, .

Based on the NYT, OpenAI used its Whisper speech recognition software to transcribe a couple of million hours of YouTube movies, which have been then used to coach GPT-4. beforehand reported that OpenAI had used YouTube movies and podcasts to coach the 2 AI programs. OpenAI president Greg Brockman was reportedly among the many folks on this crew. Per Google’s guidelines, “unauthorized scraping or downloading of YouTube content material” just isn’t allowed, Matt Bryant, a spokesperson for Google, instructed NYT, additionally saying that the corporate was unaware of any such use by OpenAI.

The report, nonetheless, claims there have been folks at Google who knew however didn’t take motion in opposition to OpenAI as a result of Google was utilizing YouTube movies to coach its personal AI fashions. Google instructed NYT it solely does so with movies from creators who’ve agreed to participate in an experimental program. Engadget has reached out to Google and OpenAI for remark.

The NYT report additionally claims Google tweaked its privateness coverage in June 2022 to extra broadly cowl its use of publicly accessible content material, together with Google Docs and Google Sheets, to coach its AI fashions and merchandise. Bryant instructed NYT that that is solely finished with the permission of customers who decide into Google’s experimental options, and that the corporate “didn’t begin coaching on extra kinds of knowledge primarily based on this language change.”