Reply to thread

Message

<blockquote data-quote="talien" data-source="post: 9467606" data-attributes="member: 3285">That's really the question, isn't it? Everything I'm seeing is it's an edge case. Certainly, newspapers have been threatening legal action against OpenAI (of ChatGPT fame) and OpenAI is either settling or making financial deals in advance for access. That said, I'm not hearing how journalists are being compensated for the rights to their articles being sold, and maybe they have no legal leg to stand on. It's a bit like how Reddit &quot;sold&quot; rights to its forums to Google so Gemini could train on it. I guess that means if you contributed to Reddit your dialogue got sold to Google (and presumably, buried in the agreement to use the site, is rights granting them the ability to do what they want with your data): <a href="https://www.reuters.com/technology/reddit-ai-content-licensing-deal-with-google-sources-say-2024-02-22/" target="_blank">https://www.reuters.com/technology/reddit-ai-content-licensing-deal-with-google-sources-say-2024-02-22/</a>The law is fuzzy on a lot of this stuff but I imagine these sorts of CD-ROM/archiving battles set a sort of precedent, and I don't see AI-data farming tremendously different. It's just creating an archive for a computer vs. for human eyes, and the subsequent possible loss of income.</blockquote>

[QUOTE="talien, post: 9467606, member: 3285"] That's really the question, isn't it? Everything I'm seeing is it's an edge case. Certainly, newspapers have been threatening legal action against OpenAI (of ChatGPT fame) and OpenAI is either settling or making financial deals in advance for access. That said, I'm not hearing how journalists are being compensated for the rights to their articles being sold, and maybe they have no legal leg to stand on. It's a bit like how Reddit "sold" rights to its forums to Google so Gemini could train on it. I guess that means if you contributed to Reddit your dialogue got sold to Google (and presumably, buried in the agreement to use the site, is rights granting them the ability to do what they want with your data): [URL]https://www.reuters.com/technology/reddit-ai-content-licensing-deal-with-google-sources-say-2024-02-22/[/URL] The law is fuzzy on a lot of this stuff but I imagine these sorts of CD-ROM/archiving battles set a sort of precedent, and I don't see AI-data farming tremendously different. It's just creating an archive for a computer vs. for human eyes, and the subsequent possible loss of income. [/QUOTE]

Verification