Press

In an interview with Bloomberg Surveillance, professor of the practice Gary Gensler said that there are safety issues, pointing to when OpenAI models breached the technology firm Hugging Face earlier this year. He noted concerns around AI's capabilities and said that while it is possible to put up safeguards, competition with China presents a "big, consequential wrinkle."

Associate dean Christopher Knittel and co-author wrote: "A simple accounting check shows why our estimated magnitude isn't a stretch. In Massachusetts, fixed costs make up roughly 60 percent of total system costs, and about 13 percent of customers have rooftop solar and pay very little, if anything, toward those costs because net metering zeroes out their bills. When this happens, the fixed cost of running the grid doesn't shrink; it just gets divided over a base that's 13 percent smaller."

Institute Professor Daron Acemoglu wrote: "The problem is not that models are too advanced (I do not see any compelling evidence that models will escape human control if they are better trained and monitored). It is that frontier labs are training their models in ways that may be leading to a type of distorted intelligence. Ultimately, it is the race for artificial general intelligence, combined with faulty metrics and hasty choices on safety, that is creating distorted intelligence and increasingly dangerous AI models."

Senior lecturer Paul Cheek said the first step for Chief Human Resources Officers (CHROs) is investing in AI training and upskilling for executives and employees. The next is rethinking the org chart. "The CHRO title itself needs to change from managing headcount to auditing whether a given role needs a human's judgment or an AI agent's uptime," Cheek said.

A paper by associate dean Christopher Knittel and co-authors argued that the administration's calculations contained major inconsistencies, such as in its assumptions about how consumers value the benefits of saving fuel. "If you do the analysis right, we found that revoking these standards does not save consumers money," said Knittel.

Assistant professor Chara Podimata is collaborating with MIT professors Adam Berinsky and Charles Stewart III on a public experiment, the Election LLM Observatory: a dashboard where the responses of major models to questions about candidates and elections are published and can be examined by journalists, researchers, and ordinary users. "I feel like we haven't taken the right steps to educate people, to make them understand what these chatbots can and can't do," she said. "Technology is changing the world very drastically and very quickly. We have left people behind. We have given them the technology itself, fully accessible, without educating them on how to use it."