Gabriele Farina grew up in a small town in the mountainous wine-growing region of northern Italy. Neither of his parents had a college degree, and although both were convinced that he “didn’t understand math”, Farina says, they bought him the technical books he wanted and did not discourage him from attending a science-oriented, rather than a classical, high school.
By the age of about 14, Farina had focused on an idea that would prove fundamental to his career.
“I was fascinated very early on by the idea that a machine could make much better predictions or decisions than humans could,” he says. “The fact that human-made mathematics and algorithms can create systems that, in some senses, outperform their creators while building on simple building blocks has always been a major source of awe for me.”
At age 16, Farina wrote code to solve a board game he played with his 13-year-old sister.
“I used game after game to calculate the optimal move and prove to my sister that she had already lost before either of us saw it,” says Farina, adding that her sister was less thrilled with her new system.
Now an assistant professor in MIT’s Department of Electrical Engineering and Computer Science (EECS) and a principal investigator in the Laboratory for Information and Decision Systems (LIDS), Farina combines concepts of game theory with tools such as machine learning, optimization, and statistics to advance the theoretical and algorithmic foundations for decision making.
Enrolling at the Politecnico di Milano for college, Farina studied automation and control engineering. However, over time, he realized that what activated his interest was “not just applying known techniques, but understanding their foundations and expanding on them,” he says. “I gradually shifted more and more toward theory, while still caring deeply about demonstrating concrete applications of that theory.”
Farina’s advisor at the Politecnico di Milano, Professor and Researcher in Computer Science and Engineering Nicola Gatti, introduced Farina to research questions in computational game theory and encouraged him to apply for a PhD. At the time, being the first in his family to earn a college degree and living in Italy, where doctoral degrees are handled differently, Farina says he didn’t even know what a PhD was.
Nevertheless, a month after graduating with his bachelor’s degree, Farina began a doctoral degree in computer science at Carnegie Mellon University. There, he achieved distinction for his research and dissertation, as well as a Facebook Fellowship in Economics and Computation.
While he was completing his doctoral studies, Farina worked for a year as a research scientist in META’s Fundamental AI Research Labs. One of his major projects was to help develop Cicero, an AI capable of defeating human players in games that involved forming alliances, negotiating, and detecting when other players were cheating.
“When we built Cicero, we designed it so that he wouldn’t agree to form an alliance if it wasn’t in his best interest, and it also understood whether a player was possibly lying, because it would be against their own incentives for them to do as they proposed,” says Farina.
An article from 2022 in MIT Technology Review That said, Cicero could represent progress toward AI that can solve complex problems that require compromise.
After his year at META, Farina joined the MIT faculty. In 2025, he was awarded a National Science Foundation Career Award. Their work – based on game theory and its mathematical language that explains what happens when different parties have different objectives, and then quantifying the “equilibrium” where no one has any reason to change their strategy – is aimed at simplifying large-scale, complex real-world scenarios where such an equilibrium could take a billion years to calculate.
“I research how we can use optimization and algorithms to actually find these stable points efficiently,” he says. “Our work seeks to shed new light on the mathematical underpinnings of the theory, try to better control and predict these complex dynamical systems, and use these ideas to compute good solutions for large multi-agent interactions.”
Farina is particularly interested in settings with “imperfect information”, meaning that some agents have information that is unknown to other participants. In such scenarios, information has value, and participants must be strategic about acting on the information they have so as not to reveal it and reduce its value. An everyday example occurs in the game of poker, where players bluff to hide information about their cards.
According to Farina, “We now live in a world in which machines are far better at deception than humans.”
The situation with “vast amounts of incomplete information” has brought Farina back to his board-game beginnings. Stratego is a military strategy game that has inspired research efforts costing millions of dollars to create systems capable of defeating human players. Farina says that requiring complex risk calculations and misdirection, or bluffing, it was probably the only classical game for which major efforts failed to produce superhuman performance.
With the new algorithm and training costing less than $10,000 instead of millions, Farina and his research team were able to defeat the best player of all time with 15 wins, four draws, and one loss. Farina says he’s thrilled to be delivering such results economically, and he hopes “these new technologies will be incorporated into future pipelines,” he says.
“We have seen continued progress toward building algorithms that can reason strategically and make sound decisions despite large task spaces or incomplete information. I am excited to see these algorithms join the broader AI revolution happening around us.”
