Over the last few weeks, I've been writing about an interesting disconnect in how we use Large Language Models (LLMs): They offer tremendous potential to transform our work - yet the measured personal productivity gains so far have been modest, often just 3-5%, or 10-20 minutes saved per day.