Post Ade0pyvDwkrOxDFJTs by palewire@mastodon.palewi.re
 (DIR) More posts by palewire@mastodon.palewi.re
 (DIR) Post #Ade0pyvDwkrOxDFJTs by palewire@mastodon.palewi.re
       2024-01-08T17:34:36Z
       
       0 likes, 1 repeats
       
       More than half of surveyed news publishers now opt-out of OpenAI's ChatGPT.That's according to the latest data gathered by our open-source homepages.news dashboard, which has been cited by the FTC, WaPo, TechCrunch, Reuters Institute and others.https://palewi.re/docs/news-homepages/openai-gptbot-robotstxt.html
       
 (DIR) Post #Ade4csI7ghG0RQxjWa by repeattofade@tootr.co
       2024-01-08T18:39:07Z
       
       0 likes, 0 repeats
       
       @palewire @ken i assume the robots exclusions only count *from that point forward*, right?  like they won't remove already crawled and ingested content from the LLMs.  basically, chatGPT is not going to unlearn content that it's publishers indicate they don't want copied?
       
 (DIR) Post #Ade4ctiOOVrerCGE6q by ken@toots.schwanksta.com
       2024-01-08T19:14:57Z
       
       0 likes, 0 repeats
       
       @repeattofade  I do not know, but I doubt they would unlearn. @palewire might know!
       
 (DIR) Post #Ade6XWaZ9KofaoAKXY by palewire@mastodon.palewi.re
       2024-01-08T19:36:21Z
       
       0 likes, 0 repeats
       
       @ken @repeattofade The OpenAI documentation for the robots.txt rules are unclear on this questionhttps://platform.openai.com/docs/gptbotOpenAI just wrote a letter responding to the NYT lawsuit. I don't believe it addresses this question directly.https://openai.com/blog/openai-and-journalism