The Risk of ‘Good Enough’ in Large Language Models

In our discussion about large language models and other AI systems’ adequacy, “good enough” should not be an endpoint. It’s where we begin.