Do you pay for the opensource projects which the "AI" companies used to train your LLMs?

Used them first to train LLMs to earn money, then exploit them again?