OpenAI and Google trained their AI models on text transcribed from YouTube videos, potentially violating creators' copyrights. The report highlights the extensive efforts made by these companies to maximize data input for AI training, with OpenAI reportedly transcribing over a million hours of YouTube videos.
Google, in response to claims of unauthorized scraping of YouTube content by OpenAI, stated this was against their rules and mentioned their unawareness of such use. However, the report suggested that despite this knowledge at Google, no action was taken, possibly due to Google's own use of YouTube content for AI model training.
The New York Times report also mentioned Google's broadened privacy policy to encompass the use of publicly available content, such as Google Docs and Sheets, for training AI models. Google clarified that this was only done with user consent, and the language update did not lead to immediate training on new types of data.
[
add
]
[
|
|
...
]