MIT Sloan School of Management

Employees using generative AI fall into one of three categories: cyborgs, who collaborate closely with artificial intelligence; centaurs, who engage with AI in a more constrained way; and self-automators, who offload a task more fully to AI. bit.ly/4rWEWdS

Agentic artificial intelligence is a promising approach for analyzing health data to improve patient outcomes, but new research shows a gap between where organizations focus their attention and where success with AI agents is determined. bit.ly/40uxOda

Companies should be proactive in managing generative AI deployment, not only to get the results that they want but to train the employees in the skills they need. bit.ly/4rWEWdS

Data deserts are markets where limited, fragmented, or low-quality data makes it difficult for AI systems to produce accurate predictions. They can distort market visibility and limit how organizations identify and reach certain groups. bit.ly/485wh1p

When aircraft noise rises by 1 decibel, on average nearby homes shed 0.6% to 1% of their value, according to a new study co-authored by Christopher Knittel, faculty director of @MIT_CPC and a professor of energy economics at MIT Sloan. bit.ly/4lG9zCJ

To shrink their AI risk exposure, organizations should inventory every instance of generative AI in use, manage embedded and enacted risks, and assign clear ownership of ongoing risk assessment. bit.ly/3Nlr6n6

Artificial intelligence has dominated business attention for several years. But the hype cycle is slowing as organizations confront the challenges of enterprise AI deployment and the need to drive tangible business value. bit.ly/4uc5jPo

The new Question Burst Catalyst tool from MIT Sloan’s Hal Gregersen helps leaders use large language models as a thought partner to help surface ideas that truly matter. bit.ly/4b1dQNx

As AI shortens the learning curve, employers must ensure that workers are using it to learn rather than as a replacement for foundational skills. bit.ly/3N6EMlL

“It is a privilege, in my opinion, to work with exceptional young people like the ones we have at MIT, in ability and character and aspiration. They actually make me a better scientist, and a better person.” — MIT Sloan professor Dimitris Bertsimas news.mit.edu/2026/algorithm…

Artificial intelligence increasingly shapes who gets a loan, which job candidates are shortlisted, and which small businesses surface in online marketplaces. But what happens when certain people simply don’t show up in the data? bit.ly/485wh1p

Today’s flight paths are more direct, which cuts fuel use and reduces emissions, but they may concentrate noise in some neighborhoods. bit.ly/4lG9zCJ

Private-sector data can be used to complement official statistics but can’t serve as a full substitute. bit.ly/3Nszhhn

There are two types of generative AI risk: embedded risks, which are inherent to the technology, and enacted risks, which are related to how it is deployed by organizations. bit.ly/3Nlr6n6

Data deserts are markets where limited, fragmented, or low-quality data makes it difficult for AI systems to produce accurate predictions. They can distort market visibility and limit how organizations identify and reach certain groups. bit.ly/485wh1p

Stagnant productivity, a tightening workforce, and rising global competition are reprioritizing manufacturing in the U.S. Four trends will determine what kind of production model will be competitive in this environment. bit.ly/4aUmSvo

A few years ago, generative artificial intelligence was barely on the public radar. Now, attention has shifted to the next evolution of generative AI: agentic AI, a new breed of AI systems that are able to perceive, reason, and act on their own. bit.ly/4c1Gkri

Noise exposure of residential properties affects property prices. Future airport design and urban planning decisions should take those effects into account. bit.ly/4lG9zCJ

Risk-averse technology investments, wage stagnation, and the U.S. tariff strategy are just some of the factors impeding expected productivity gains. bit.ly/4tQwqPv

Research shows that the hardest work in deploying agentic AI in a clinical setting is the “sociotechnical aspects” — rather than tasks like prompt engineering. bit.ly/40uxOda