Billy Luedtke, founder of Intuition, has raised concerns about the quality of data used to train AI models. He describes the current situation as 'slop-in, slop-out,' indicating that AI systems are increasingly relying on polluted data from the open web, which results in distorted outputs. Luedtke advocates for the implementation of a trust layer in AI, which would include verifiable attribution and decentralized reputation systems to improve data quality and reliability.