MIT | Guided learning lets “untrainable” neural networks realize their potential
Dec 18, 2025 - MIT CSAIL study suggests that neural network architectures considered unsuitable for modern tasks can improve with short-term guidance.
Deep-learning model predicts how fruit flies form, cell by cell
Dec 15, 2025 - The method may help scientists predict the development of more complex tissues or identify early signs of diseases such as asthma and cancer.
MIT study finds AI can already replace 11.7% of U.S. workforce
Nov 26, 2025 - Artificial intelligence can already replace 11.7% of the U.S. labor market, across finance, health care and professional services, according to MIT's study.
Understanding the nuances of human-like intelligence
Nov 11, 2025 - Can we better understand ourselves if we better understand the artificial intelligence systems that are becoming a more significant part of our everyday lives?
Scaling Laws: AI and Energy: What do we know? What are we learning?
Oct 14, 2024 - Discussing the energy costs of artificial intelligence.
AI and science intersect at Michigan AI Symposium
Oct 27, 2025 - The event brought together over 300 researchers and practitioners to explore how artificial intelligence is driving scientific advances across disciplines.
Charts can be social artifacts that communicate more than just data
Oct 22, 2025 - The degree to which someone trusts the information depicted in a chart can depend on their assumptions about who made the data visualization, according to a pair of studies by MIT researchers.
Amazon launches $68 million AI PhD Fellowship program
Oct 21, 2025 - The initiative will fund over 100 doctoral students researching machine learning, computer vision, and natural-language processing at nine universities.
Uchicago DSI - Why Can’t Powerful LLMs Learn Multiplication?
Oct 24, 2025 - This paper’s key contribution is architectural insights and training techniques can overcome obstacles that scaling alone cannot address.
MIT | Method teaches generative AI models to locate personalized objects
Oct 16, 2025 - Researchers from MIT and the MIT-IBM Watson AI Lab have introduced a new training method that teaches vision-language models to localize personalized objects in a scene.
MIT | Ray Kurzweil ’70 reinforces his optimism in tech progress
Oct 10, 2025 - Innovator, futurist, author, and MIT alumnus Ray Kurzweil emphasized his optimism about AI and technological progress in a lecture at MIT while accepting the Robert A. Muh Alumni Award from the School of Humanities, Arts, and Social Sciences.
Using generative AI to diversify virtual training grounds for robots
Oct 8, 2025 - The current amount of data aren’t enough to teach a robot to be a helpful household or factory assistant.
MIT | New prediction model could improve the reliability of fusion power plants
Oct 7, 2025 - These fusion machines use powerful magnets to contain a plasma hotter than the sun’s core and push the plasma’s atoms to fuse and release energy.
MIT Schwarzman College of Computing and MBZUAI launch international collaboration to shape the future of AI
Oct 3, 2025 - The new effort to strengthen the building blocks of artificial intelligence and accelerate its use in pressing scientific and societal challenges.
AI maps how a new antibiotic targets gut bacteria
Oct 3, 2025 - Researchers used a generative AI model to reveal how a narrow-spectrum antibiotic attacks disease-causing bacteria, speeding up a process that normally takes years.
MIT | Responding to the climate impact of generative AI
Sep 30, 2025 - MIT experts discuss strategies and innovations aimed at mitigating the amount of greenhouse gas emissions generated by the training, deployment, and use of AI systems, in the second in a two-part series on the environmental impacts of generative artificial intelligence.
MIT AI system learns from many types of scientific information and runs experiments to discover new materials
Sep 25, 2025 - . The system could generate solutions to energy problems that have plagued the materials science and engineering community for decades.
New tool makes generative AI models more likely to create breakthrough materials
Sep 22, 2025 - A new tool called SCIGEN allows researchers to implement design rules that AI models must follow when generating new materials. The advance could speed the development of materials that enable technological breakthroughs.
How to build AI scaling laws for efficient LLM training and budget maximization
Sep 16, 2025 - MIT and MIT-IBM Watson AI Lab researchers have developed a universal guide for estimating how large language models (LLMs) will perform based on smaller models in the same family.
A human-centered approach to data visualization
Sep 5, 2025 - MIT Associate Professor Arvind Satyanarayan develops tools for data visualization that amplify creativity and promote human agency.
AI and machine learning for engineering design
Sep 7, 2025 - In MIT course 2.155/156 (AI and Machine Learning for Engineering Design), students use tools and techniques from artificial intelligence and machine learning for mechanical engineering design, focusing on the creation of new products and addressing engineering design challenges.
95% of organizations got zero return on AI investment in MIT study
Aug 21, 2025 - Companies are investing tens of billions of dollars in generative AI with little to show for it.
A new model predicts how molecules will dissolve in different solvents
Aug 19, 2025 - This type of prediction could make it much easier to develop new ways to produce pharmaceuticals and other useful molecules.
LinkedIn Top Colleges 2025: The 50 best colleges for long-term career success in the U.S.
Aug 12, 2025 - See the 2025 LinkedIn Top Colleges list, a data-backed ranking of the 50 best colleges for career success in the U.S.
AI is already displacing these jobs
Aug 18, 2025 - Any disruption from AI is so far hitting farther afield, the MIT findings suggest.
Helping data storage keep up with the AI revolution
Aug 4, 2025 - Cloudian, co-founded by MIT alumnus Michael Tso, has created a storage system to help businesses feed data-hungry AI models and agents at scale.
Where is all the NIH data? A Landscape Analysis of the 150 NIH Repositories and the Path Beyond the Mountains
Jul 31, 2025 - 'Less is more' when establishing repositories that help scientists easily find and securely access distinct data types for their research. This level of cross architecture coordination and federation will require support from both research grantors and recipients.
A new way to edit or generate images
Jul 21, 2025 - MIT and Meta researchers discover that special kinds of neural networks called encoders or “tokenizers” can do much more than previously realized, leading to a new way to edit or generate images.
Unpacking the bias of large language models
Jun 17, 2025 - MIT researchers discovered the underlying cause of position bias, a phenomenon that causes large language models to overemphasize the beginning or end of a document or conversation, while neglecting the middle.
Meet the MIT engineer who invented an AI-powered way to restore art
Jun 11, 2025 - The new method could be 70 times faster than repairing painting by hand.