AI Flash News

Helping AI agents search to get the best results out of large language models Published: Thu, 05 Feb 2026 16:30:00 -0500 | (Link) EnCompass executes AI agent programs by backtracking and making multiple attempts, finding the best set of outputs generated by an LLM. It could help coders work with AI agents more efficiently. Brian […]

Why it’s critical to move beyond overly aggregated machine-learning metrics Published: Tue, 20 Jan 2026 16:30:00 -0500 | (Link) New research detects hidden evidence of mistaken correlations — and provides a method to improve accuracy. At MIT, a continued commitment to understanding intelligence Published: Wed, 14 Jan 2026 16:50:00 -0500 | (Link) With support from […]

At MIT, a continued commitment to understanding intelligence Published: Wed, 14 Jan 2026 16:50:00 -0500 | (Link) With support from the Siegel Family Endowment, the newly renamed MIT Siegel Family Quest for Intelligence investigates how brains produce intelligence and how it can be replicated to solve problems. Generative AI tool helps 3D print personal items […]

3 Questions: How AI could optimize the power grid Published: Fri, 09 Jan 2026 00:00:00 -0500 | (Link) While the growing energy demands of AI are worrying, some techniques can also help make power grids cleaner and more efficient. Decoding the Arctic to predict winter weather Published: Thu, 08 Jan 2026 16:55:00 -0500 | (Link) […]

MIT in the media: 2025 in review Published: Mon, 22 Dec 2025 17:20:00 -0500 | (Link) MIT community members made headlines with key research advances and their efforts to tackle pressing challenges. Guided learning lets “untrainable” neural networks realize their potential Published: Thu, 18 Dec 2025 16:20:00 -0500 | (Link) CSAIL researchers find even “untrainable” […]

MIT in the media: 2025 in review Published: Mon, 22 Dec 2025 17:20:00 -0500 | (Link) MIT community members made headlines with key research advances and their efforts to tackle pressing challenges. Guided learning lets “untrainable” neural networks realize their potential Published: Thu, 18 Dec 2025 16:20:00 -0500 | (Link) CSAIL researchers find even “untrainable” […]

Guided learning lets “untrainable” neural networks realize their potential Published: Thu, 18 Dec 2025 16:20:00 -0500 | (Link) CSAIL researchers find even “untrainable” neural nets can learn effectively when guided by another network’s built-in biases using their guidance method. A new way to increase the capabilities of large language models Published: Wed, 17 Dec 2025 […]

New method improves the reliability of statistical estimations Published: Fri, 12 Dec 2025 00:00:00 -0500 | (Link) The technique can help scientists in economics, public health, and other fields understand whether to trust the results of their experiments. New materials could boost the energy efficiency of microelectronics Published: Thu, 11 Dec 2025 00:00:00 -0500 | […]

MIT researchers “speak objects into existence” using AI and robotics Published: Fri, 05 Dec 2025 10:00:00 -0500 | (Link) The speech-to-reality system combines 3D generative AI and robotic assembly to create objects on demand. Robots that spare warehouse workers the heavy lifting Published: Fri, 05 Dec 2025 00:00:00 -0500 | (Link) Founded by MIT alumni, […]

Researchers discover a shortcoming that makes LLMs less reliable Published: Wed, 26 Nov 2025 00:00:00 -0500 | (Link) Large language models can learn to mistakenly link certain sentence patterns with specific topics — and may then repeat these patterns instead of reasoning. MIT scientists debut a generative AI model that could create molecules addressing hard-to-treat […]

The cost of thinking Published: Wed, 19 Nov 2025 16:45:00 -0500 | (Link) MIT neuroscientists find a surprising parallel in the ways humans and new AI models solve complex problems. New AI agent learns to use CAD to create 3D objects from sketches Published: Wed, 19 Nov 2025 00:00:00 -0500 | (Link) The virtual VideoCAD […]

Understanding the nuances of human-like intelligence Published: Tue, 11 Nov 2025 00:00:00 -0500 | (Link) Associate Professor Phillip Isola studies the ways in which intelligent machines “think,” in an effort to safely integrate AI into human society. MIT Energy Initiative launches Data Center Power Forum Published: Fri, 07 Nov 2025 14:55:00 -0500 | (Link) MIT […]

Charting the future of AI, from safer answers to faster thinking Published: Thu, 06 Nov 2025 16:40:00 -0500 | (Link) MIT PhD students who interned with the MIT-IBM Watson AI Lab Summer Program are pushing AI tools to be more flexible, efficient, and grounded in truth. MIT researchers propose a new model for legible, modular […]

The brain power behind sustainable AI Published: Fri, 24 Oct 2025 00:00:00 -0400 | (Link) PhD student Miranda Schwacke explores how computing inspired by the human brain can fuel energy-efficient artificial intelligence. Five with MIT ties elected to National Academy of Medicine for 2025 Published: Wed, 22 Oct 2025 15:25:00 -0400 | (Link) Professors Facundo […]

The brain power behind sustainable AI Published: Fri, 24 Oct 2025 00:00:00 -0400 | (Link) PhD student Miranda Schwacke explores how computing inspired by the human brain can fuel energy-efficient artificial intelligence. Five with MIT ties elected to National Academy of Medicine for 2025 Published: Wed, 22 Oct 2025 15:25:00 -0400 | (Link) Professors Facundo […]

Method teaches generative AI models to locate personalized objects Published: Thu, 16 Oct 2025 00:00:00 -0400 | (Link) After being trained with this technique, vision-language models can better identify a unique item in a new scene. Remembering Professor Emerita Jeanne Shapiro  Bamberger, a pioneer in music education Published: Wed, 15 Oct 2025 13:25:00 -0400 | […]