They claim it’s “clean room”, so no decompilation involved. I’m just waiting for the lawsuits to arrive, so we may learn how much of that is true. The big ones probably are already checking the code.
Companies were just dropping the entirety of their proprietary code base into chatbots to save money after firing their best developers. I bet that’s where they’re getting most of it.
I think that could only be really claimed if you know what training data the LLM was trained on. That will only matter if it ends up in court though and no one seems willing to fight that fight at the moment.
They claim it’s “clean room”, so no decompilation involved. I’m just waiting for the lawsuits to arrive, so we may learn how much of that is true. The big ones probably are already checking the code.
Companies were just dropping the entirety of their proprietary code base into chatbots to save money after firing their best developers. I bet that’s where they’re getting most of it.
I think that could only be really claimed if you know what training data the LLM was trained on. That will only matter if it ends up in court though and no one seems willing to fight that fight at the moment.