--- title: "Teaching AI models to say “I’m not sure”" url: https://daily.dev/posts/teaching-ai-models-to-say-i-m-not-sure--kwyridj4a source_url: https://news.mit.edu/2026/teaching-ai-models-to-say-im-not-sure-0422 type: article source: "MIT News" published: 2026-04-22T19:25:51.034Z updated: 2026-04-22T19:26:10.228Z tags: ["llm", "reinforcement-learning"] reading_time: 4 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Teaching AI models to say “I’m not sure” **[MIT News](https://daily.dev/sources/mit)** · 4 min read · 0 upvotes · 0 comments ## Summary MIT researchers have developed a training technique called 'reinforcement learning with calibration rewards' (or 'reinforcement learning with calibration rewards') that teaches AI reasoning models to express calibrated confidence alongside their answers. Standard reinforcement learning training rewards correct answers but ignores uncertainty, causing models to become overconfident even when guessing. By adding a 'brier score' term to the reward function, the new approach penalizes the gap between stated confidence and actual accuracy. In experiments on a 7-billion-parameter model, the technique reduced calibration error by up to 90% without sacrificing accuracy, and generalized to unseen benchmarks. The method also outperformed post-hoc confidence classifiers and proved useful at inference time for selecting or weighting candidate answers. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [The AI Model Confidence Trap](https://daily.dev/posts/the-ai-model-confidence-trap-93to0xhgy) · Towards Data Science · 5 upvotes · 1 comments - [A smarter way for large language models to think about hard problems](https://daily.dev/posts/a-smarter-way-for-large-language-models-to-think-about-hard-problems-d6aqwfq0t) · MIT News · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/teaching-ai-models-to-say-i-m-not-sure--kwyridj4a)