Computer Science and Artificial Intelligence Laboratory (CSAIL)

Prof. Andrew Lo speaks with Bloomberg reporter Lu Wang about how AI tools could be applied to the financial services industry, working alongside humans to help manage money, balance risk, tailor strategies and possibly even act in a client’s best interest. “I believe that within the next five years we’re going to see a revolution in how humans interact with AI,” says Lo. He adds that “the financial services industry has extra layers of protection that needs to be built before these tools can be useful.”

Researchers at MIT have uncovered a variety of obstacles of AI in software development, reports Rob Wile for NBC News. They have found “the main obstacles come when AI programs are asked to develop code at scale, or with more complex logic,” writes Wile. 

A new research paper by Prof. David Autor and Principal Research Scientist Neil Thompson explores the forthcoming impact of AI on jobs, reports Tim Harford for Financial Times. “[W]hile there are few certainties, Autor and Thompson’s framework does suggest a clarifying question: does AI look like it is going to do the most highly skilled part of your job or the low-skill rump that you’ve not been able to get rid of?,” writes Harford. “The answer to that question may help to predict whether your job is about to get more fun or more annoying — and whether your salary is likely to rise, or fall as your expert work is devalued like the expert work of the Luddites.” 

Forbes contributor Tanya Fileva spotlights how MIT CSAIL researchers have developed a system called Air-Guardian, an “AI-enabled copilot that monitors a pilot’s gaze and intervenes when their attention is lacking.” Fileva notes that “in tests, the system ‘reduced the risk level of flights and increased the success rate of navigating to target points’—demonstrating how AI copilots can enhance safety by assisting with real-time decision-making.”

Prof. Ryan Williams has published a new proof that explores computational complexity and flips the script on years of assumptions about the trade-offs between computation space and time, reports Max Springer for Scientific American. Williams found that “any problem can be transformed into one you can solve by cleverly reusing space, deftly cramming the necessary information into just a square-root number of bits,” Springer explains. “This progress is unbelievable,” says Mahdi Cheraghchi of the University of Michigan. “Before this result, there were problems you could solve in a certain amount of time, but many thought you couldn’t do so with such little space.” 

Prof. Daniela Rus, director of CSAIL, speaks with Newsweek reporter Marni Rose McFall about the impact of AI on entry level jobs. “We need a strong pipeline of talent that starts with entry-level roles, internships, and hands-on learning opportunities," says Rus. "These early experiences remain essential stepping stones, helping people build technical confidence, domain fluency, and problem-solving skills. And soon, the skills companies will be looking for in entry-level workers is how well they can make the most of AI tools."

Prof. Daniela Rus, director of CSAIL, and research affiliate Ramin Hasani have been named to The Boston Globe’s 2025 list of Tech Power Players working in the foundational AI sector, reports Aaron Pressman for The Boston Globe. Rus and Hasani are co-founders of Liquid AI, a startup that has developed “an AI technique with fewer software ‘neurons’ than large language models of OpenAI and others,” explains Pressman. This means “Liquid AI requires less computing power (and electricity.)” 

MIT researchers have developed Boltz-2, an AI algorithm “that unites protein folding and prediction of small-molecule binding affinity in one package,” reports Laura Howes for Chemical & Engineering News. “The researchers say their new AI model approaches the level of accuracy achieved by traditional computational chemistry—such as methods involving free-energy perturbation calculations—but much more quickly and cheaply,” explains Howes. 

Researchers at MIT and elsewhere have developed Boltz-2, an open-source generative AI model that can help researchers find new medicines faster, reports Alex Knapp for Forbes. The tool “can not only predict the structure of proteins, it can also predict its binding affinity–that is, how well a potential drug is able to interact with that protein,” explains Knapp. “This is crucial in the early stages of developing a new medicine.” 

Wall Street Journal reporter Angelina Torre spotlights “Letterlocking: The Hidden History of the Letter,” a new book by MIT Libraries Conservator Jana Dambrogio and King’s College London Senior Lecturer Daniel Smith that explores the history and art of “folding a letter so it serves as its own envelope.” The book “calls on scholars to ‘read the folds’ of written correspondence – to peer into the historical, social or personal circumstances that might not be explicitly stated,” explains Torre. 

Prof. Daniela Rus, director of CSAIL, speaks with Financial Times reporter Michael Peel about ongoing efforts to balance autonomous vehicles’ “efficient operation with the need for them to minimize damage in collisions.” Rus notes that a new framework offers a “potential path towards AVs that can navigate complex, multi-agent scenarios with an awareness of differing levels of vulnerability among road users,” says Rus.

Prof. Daniela Rus, director of CSAIL, speaks with Wall Street Journal reporter Isabelle Bousquette about her vision for the future of robots as soft, squishy, flexible and maybe even edible. Bousquette notes that Rus is a “pioneer” in the field of soft robotics and Steve Crowe, chair of the Robotics Summit and Expo, emphasizes: “there’s literally nobody in the world that knows more about this stuff than Daniela Rus.” “I really wanted to broaden our view of what a robot is,” says Rus. “If you have a mechanism that’s made out of paper and that moves, is that a robot or not? If you have an origami flower that you attach to a motor, is that a robot or not? To me, it’s a robot.” 

CSAIL Research Scientist Neil Thompson speaks with Wired reporter Will Knight about how new AI systems are developing new algorithms that could be used to help advance scientific research and innovation. “If these capabilities can be used to tackle bigger, less tightly-scoped problems, it has the potential to accelerate innovation—and thus prosperity,” says Thompson.

Prof. Pulkit Agrawal speaks with Darian Woods and Geoff Brumfiel of NPR’s The Indicator from Planet Money about his work developing a simulator that can be used to train robots. “The power of simulation is that you can collect, you know, very large amounts of data,” explains Agrawal. “For example, in three hours', you know, worth of simulation, we can collect 100 days' worth of data.” 

Researchers at MIT have concluded that AI does not develop “value systems” over time, reports Kyle Wiggers for TechCrunch. “For me, my biggest takeaway from doing all this research is to now have an understanding of models as not really being systems that have some sort of stable, coherent set of beliefs and preferences,” says graduate student Stephen Casper. “Instead, they are imitators deep down who do all sorts of confabulation and say all sorts of frivolous things.”