Publishers and software vendors rewrote their terms around AI training this month. The New York Times closed its content to AI training twice, first in its Terms of Service and then in robots.txt. Zoom revised the same clause twice in five days after public criticism, ending on a flat statement that it does not train AI models on meeting content.

Zoom published new Terms of Service effective August 11, 2023. The consent caveat was replaced with a flat statement: "Zoom does not use any of your audio, video, chat, screen sharing, attachments or other communications-like Customer Content (such as poll results, whiteboard and reactions) to train Zoom or third-party artificial intelligence models."

The New York Times updated its Terms of Service (last updated moved from December 19, 2022 to August 3, 2023). Section 2.1 now excludes using its Content without prior written consent in connection with "the development of any software program, including, but not limited to, training a machine learning or artificial intelligence (AI) system".

Zoom added a sentence to section 10.4 of its Terms of Service, effective August 7, 2023: "Notwithstanding the above, Zoom will not use audio, video or chat Customer Content to train our artificial intelligence models without your consent."

Two lines were added to robots.txt: "User-agent: GPTBot" and "Disallow: /". OpenAI's web crawler is now blocked from the whole site, two weeks after the Terms of Service change on AI training.