Stanford CS prof, human-computer interaction and social computing. Surely I will be interesting and entertaining here, even though I failed to achieve this on Twitter.
Stanford CS prof, human-computer interaction and social computing. Surely I will be interesting and entertaining here, even though I failed to achieve this on Twitter.
Despite the promise of AIs improving human decision making, a frustratingly resilient research result has found that people with AI decision-making aids are no better than people alone or AIs alone. Why? Overreliance: we go along with the AI's recommendations even when we shouldn't. Even when explainable AIs try to make it clear when the AI is wrong!
[CSCW paper thread]
Loading comments...