While KYC measures in finance are designed to prevent money laundering and terrorism financing, similar requirements for AI models could inadvertently stifle innovation and creativity. How might we balance the need for transparency with the importance of privacy and free expression in AI research? Could a more targeted approach, focusing on high-risk use cases, help mitigate these concerns while still ensuring accountability?