Confidence is persuasive. In artificial intelligence systems, it is often misleading. Today’s most capable reasoning models share a trait with the loudest voice in the room: They deliver every answer with the same unshakable certainty, whether they’re right or guessing. Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have now traced that overconfidence to a specific flaw in how these models are trained, and developed a method that fixes it without giving up any accuracy. The team’s research is published on the arXiv preprint server.
Teaching AI models to say ‘I’m not sure’ in cases of calibration errors
-
HighlightsFree Dark Web Monitoring Stamps the $17 Million Credentials Markets
-
HighlightsSmart buildings: What happens to our free will when tech makes choices for us?
-
AppsScreenshots have generated new forms of storytelling, from Twitter fan fiction to desktop film
-
HighlightsDarknet markets generate millions in revenue selling stolen personal data, supply chain study finds
-
SecurityPrivacy violations undermine the trustworthiness of the Tim Hortons brand
-
Featured HeadlinesWhy Tesla’s Autopilot crashes spurred the feds to investigate driver-assist technologies – and what that means for the future of self-driving cars
