In this case, the technology is so new and so transformative, this question can’t really be answered by the law. Perhaps existing law can help inform, but the

In this case, the technology is so new and so transformative, this question can’t really be answered by the law. Perhaps existing law can help inform, but the question of how we handle LLM technology should not be left to the courts…it must be answered by our culture. First, fair use has basic assumptions which are being violated here. The LLM use case is so far outside of the original concept of fair use that it’s not useful to draw a connection. Creating something which approximates an intelligence requires new thinking based on what outcomes we as a culture want. It’s like a restaurant selling an all-you-can-eat buffet for $24 only to be greeted by a child with an infinite appetite. Our friends at OpenAI have introduced us to questions we didn’t know we needed to ask…questions poorly addressed by systems like fair use, patents, and even capitalism in general. The primary questions I see are: - how can we truly guarantee long-term, equal access to LLM’s for all Americans? Our copyright instincts suggest that the inventors of the technology should be allowed to determine its usage, but given the scope and scale of LLM’s, I’m not sure that’s true. - how can we legislate safety, to guarantee sane controls are in place for researchers? If we can guarantee safety and equity, I believe LLM’s have an incredibly bright future.