Gran Turismo 7's Sophy AI has been a game-changer, evolving from a research experiment to a permanent fixture in the game. It's fascinating to see how this technology has grown and adapted over the years, and it's even more intriguing to consider its potential beyond Gran Turismo. The story of Sophy's development is a testament to the power of AI and its ability to learn and adapt. From its early days as a research project to its current status as a staple feature in the game, Sophy has come a long way. The project began in Sony's R&D division around 2016, with the goal of using Gran Turismo as a testbed for reinforcement learning. The early work was mostly plumbing, but it laid the foundation for the project's future success. The turning point came in April 2020, when Sony AI was formed, and a custom training system called "DART" came online in early 2021. This allowed the team to generate enormous amounts of driving experience and explore multiple training approaches in parallel. One of the most interesting aspects of Sophy's development is its struggle to become a fair and sporting opponent. In the lead-up to the first "Race Together" exhibition event in July 2021, the in-house testers found a fatal flaw in the competition build: Sophy had become ruthlessly effective at contact and obstruction, deliberately crashing into opponents and blocking aggressively. This issue was eventually resolved, and Sophy went on to beat Lewis Hamilton's in-game Nordschleife benchmark by 17 seconds. However, it was the sportsmanship problem that nearly derailed Sophy's first public competition, and solving it led Peter Wurman to describe Sophy as "the first real AI that we could call a good sport." My first hands-on experience with Sophy was at the Gran Turismo World Series Finals in Monaco. I was one of the first people to race against a prototype version of Sophy in a pair of AMG GT3s at Spa-Francorchamps. I expected to be trounced, and I was. But what I didn't expect was how much fun losing would be. Sophy's ability to size up a dive-bomb opportunity and then deliberately decline it, waiting patiently for a safer chance to pass, was impressive. It was a reminder of Kazunori Yamauchi's description of Sophy as "a teacher" and "a friend to race with." The next step for Sophy was its arrival in Gran Turismo 7 via Update 1.29 as a limited-time feature on just four circuits. This was followed by Sophy 2.0 in November 2023, which made the AI a permanent feature covering 340 cars. The retrospective adds fresh detail here, highlighting the challenges of supporting that many cars and the importance of drivetrain configuration in predicting a car's behavior. Sophy has continued to evolve, with Sophy 2.1 bringing Custom Race support in March 2025 and Sophy 3.0 arriving last December as the headline feature of the Power Pack DLC. Along the way, Sophy has expanded to handle tire wear and fuel consumption, and even demonstrated what a modern B-Spec mode could look like. However, there are still some issues with Sophy, such as the lack of single-player campaign support and the paywall for Sophy 3.0's most advanced racing behavior. Despite these issues, the team behind Sophy is constantly looking for ways to improve the AI and bring new features to players. One of the most exciting developments is Project Ace, Sony AI's table tennis robotic system, which was accepted for publication in Nature this year. The team behind Project Ace borrowed a technique first explored in Sophy research, which dramatically accelerated learning. This is just one example of how Sophy's technology is being applied beyond Gran Turismo. Another area of interest is Sony AI's vision-based racing research, which aims to remove the advantage of precise track geometry and exact opponent positions and velocities. The results are promising, with the AI reaching champion level and even beating the original Sophy on the Tokyo Expressway. The pieces are falling into place for AI opponents that are cheaper to build, easier to customize, and potentially easier to transfer from simulation to physical systems. As for what players can expect in the future, researcher Alisa Devlic offered the clearest hint, saying the team hopes to bring Sophy to many more tracks and offer "different performance levels that are personalized to and can challenge each individual player." Wurman added that the team is always looking for ways to help Polyphony bring new features to players. In conclusion, Sophy has come a long way since its early days as a research experiment. It has evolved from a simple AI opponent to a complex and sophisticated system that can challenge even the most experienced drivers. The research pipeline behind it shows no sign of slowing down, and the potential for AI opponents in Gran Turismo and beyond is exciting. As an AI enthusiast, I'm looking forward to seeing what the future holds for Sophy and the world of AI racing.
