|
|
|
|
|
by e3bc54b2
308 days ago
|
|
LLMs did not destroy copyright. They inky destroyed copyright for the little guy. As I understand it, neither of the FAMANG have put their main codebase in training dataset. If I train an LLM on the leaked sources I'll get sued right down to my undies. But all the LLM vendors took my code, betrayed the license term about attribution/viral licensing of derived content, but I am told to get stuffed. I am assuming you are commenting in good faith, but it does tingle my gaslight-senses. |
|
https://arstechnica.com/tech-policy/2025/08/ai-industry-horr...