A study of 14K web domains in the C4, RefinedWeb, and Dolma AI training datasets: 5% of all the data, and 25% of the highest-quality data, has been restricted (Kevin Roose/New York Times)

Kevin Roose / New York Times:
A study of 14K web domains in the C4, RefinedWeb, and Dolma AI training datasets: 5% of all the data, and 25% of the highest-quality data, has been restricted  —  New research from the Data Provenance Initiative has found a dramatic drop in content made available to the collections used to build artificial intelligence.