New method enables AI for safety-critical situations
The “HardFlow” algorithm could help generative AI models produce high-quality outputs that obey strict requirements when “pretty close” doesn’t cut it.
Search Stories
New method allows scientists to follow gene activity over time in the same cells
In the new method, cells package and export their RNA, enabling researchers to sequence and analyze the RNA without killing the cells.
Media Advisory: Governor Healey, MIT President Kornbluth to Kick Off Festivities at MIT Future Fest
The five-day festival will explore the future of science, technology, art and design from Wednesday, Sept. 30 to Sunday, Oct. 4
Faculty receive promotions in the School of Architecture and Planning for 2026
The range of faculty disciplines showcases the breadth of research and scholarship across the school.
From MIT to IBM, expediting AI and quantum deployment
MIT affiliates engage with the MIT-IBM Computing Research Lab to bring rigorous theory to production systems.
System helps humans predict when self-driving cars will make mistakes
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
Playing against climate risk
Climate scientist Sai Ravela is using games to help coastal communities find creative adaptations to climate change.
How an MIT research project became a global programming language
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
How MIT Sandbox has turned student ideas into $8.7 billion in global impact
Through a decade of non-dilutive funding, hands-on mentorship, and a safe space to fail fast, the MIT Sandbox Innovation Fund Program has helped students translate curiosity into real-world impact.
MIT engineers create a system for building shape-changing smart devices
Dubbed “bifur-circuits,” these interactive building blocks could be used to develop reconfigurable robotic grippers or customized assistive devices.
Generating scenarios for extreme events, without extreme data
A new algorithm learns to anticipate the unprecedented scenarios that critical infrastructure and global supply chains are least prepared for.
MIT engineers design a better controller for operating construction diggers
The new, more intuitive system could speed up the training process for excavator operators.
Mathematical framework connects biological principles to manufacturable, adaptive materials
The new framework could streamline the design process of robotic grippers or aerospace components that exhibit complex behaviors found in nature.
With a feel for physics, AI models simulate a wider range of real-world scenarios
“GeoPT” helps AI models understand the basics of physics so they can simulate how objects respond to things like wind and water more efficiently and accurately.
Personalized physical therapy: Stroke rehabilitation powered by AI
MIT engineers introduce an adaptive physical therapy system that uses generative AI to learn from physical therapists and interactively support stroke patients.