Measured comparisons of AI tools, run on the same task, on the same day, under rules written down and frozen before the first tool is opened. Every figure leads back to a dated run and a kept output.
Nothing is published here yet. The first comparison covers tools that record, transcribe and summarise meetings. The test meeting has been written and its reference answer frozen; results will appear here with their test date, and the method they are measured against is set out in the methodology article.