Data brief
We automated part of our own claim-check pipeline. Here is the error rate.
Where the automation matched a human grader, and where it still needs a second look before publishing.
Draft entry. The full analysis for this piece ships with the data pipeline; the dataset it grades against is being assembled.