IBM’s Deep Blue became the first computer to defeat a reigning world chess champion under standard tournament time controls, a major milestone in both chess and artificial intelligence history.

On February 10, 1996, Deep Blue won the first game of a six-game match against Garry Kasparov, the world champion at the time. Although Kasparov went on to win the match by scoring three wins and two draws in the remaining games, Deep Blue’s initial victory marked the first time a computer had ever beaten a world champion in a regulation-format game.

In May 1997, an upgraded version of Deep Blue faced Kasparov in a much-anticipated rematch. This time, the computer triumphed over the champion in a six-game series, winning two games, drawing three, and losing one – clinching the match 2 – 1 with 3 draws.

The upset shocked the chess world and led Kasparov to accuse IBM of cheating, suggesting human intervention during the match. IBM denied the allegations and declined a rematch. Instead, the company retired and dismantled Deep Blue, viewing the project as a completed demonstration of its computing capabilities.

The development of Deep Blue began in 1985 with the ChipTest project at Carnegie Mellon University. The system relied primarily on brute-force search, evaluating vast numbers of chess positions rather than mimicking human strategic reasoning. The 1997 version of Deep Blue ran on a massively parallel RS/6000 SP system with 96 nodes, each node powered by a 120 MHz P2SC microprocessor, enhanced by 480 special-purpose VLSI chess chips.

The chess-playing program of Deep Blue was developed in the C programming language, operating under the AIX operating system.

During its 1997 rematch with Kasparov, Deep Blue was capable of evaluating 200 million positions per second. This rate was twice as fast as its 1996 version, thus marking a significant leap in its computing capability. At that time, it held the rank of the world’s 259th most powerful supercomputer.

The Deep Blue-Kasparov matches were landmark events in the history of artificial intelligence. They showed the world that machines could outperform even the best human minds in complex tasks, sparking widespread debate about the implications of rapidly advancing technology.

Deep Blue paved the way for the development of more sophisticated AI systems that continue to push the boundaries of what machines can achieve. It set the stage for later breakthroughs like AlphaZero, which in the late 2010s mastered chess, Go, and shogi using self-play and reinforcement learning, and more broadly for the rise of large-scale AI models that now permeate language, science, and industry.