AI develops a taste for reviewing

"This is a sound Cabernet. It's very dry and a little thin in blackberry fruit, which accentuates the acidity and tannins. Drink up."

These tasting notes are the work of a reviewer who cannot sample the wine in question. They were written by a computer code developed in 2020 by a team of Dartmouth researchersDan Rockmore, professor of mathematics and computer science, former Neukom Institute fellow Allen Riddell, and Keith Carlson, Guarini '21. The algorithm, trained on hundreds of thousands of existing reviews, learned to craft original reviews of its own based on a set of product features, as well as synthesize a large volume of published reviews into a new one of its making. Convinced that the work would interest a wide audience, the team joined hands with faculty at the Tuck School of Business to reframe their code into "a market-friendly tool that can assist consumers, marketers, and professional reviewers."

You can read the Dartmouth News story on the collaboration at