But then If we agree on IP, we should not complain that openai want free access to copyrighted materials, we should use their own logic to force them to make their model open source, and free for anyone to execute on their own hardware.
They get free access to data so we should get free access to the compilation of the data. Then they can charge us for the hardware cost of running the model, but they’ll have to charge us no more than what it costs, because they will be competing with other company running the exact same model and driving the price down.
Well I agree with the concept of freeing IP but I don’t think the first step is to give such freedom to a closed, for profit, corporation. If OpenAI was indeed an open source project, then yes, but that is not the case today
If we give them free access to copyright material, what will happen is that it will be an exception for them but not for us. You know this.
