@tantramar @Cliff538 I think you'd have to prove that the LLM was trained on closed source proprietary data, and it probably wasn't. It just actually looked at the software and programmed an equivalent. Or at least that's what we're lead to beleive.