The human-in-the-loop is tired

# Summary Developers are experiencing genuine usefulness from LLM programming tools alongside destabilizing effects that risk burnout, according to Laura Summers at Pydantic. While AI has finally narrowed the gap between promise and reality compared to previous low-code solutions, the human burden of reviewing, directing, and correcting AI-generated code feels more exhausting rather than less so. Summers argues for acknowledging this duality honestly rather than dismissing concerns or promoting hype.