By Neve Chamberlain
Artificial intelligence (AI) is changing how esports players train and compete across the globe, but student athletes at Toronto Metropolitan University (TMU) say it can never replace the human creativity that makes gaming fun.
TMU esports fields competitive teams for several games, including Valorant, Smash Bros. and Rocket League.
“Sports is an art, right? And if you turn an art into an algorithm, it ceases to be art,” said Mateusz Sobolewski, a fifth-year computer engineering student and manager for TMU esports’ Rocket League team.
For Sobolewski and other members of TMU’s esports community, AI presents both opportunities and challenges, from improving player performance to threatening the integrity of competition.
There is a place for AI in esports, “but it’s kind of about finding the balance. That is the most important part,” said Justin Mandel, one of TMU esports’ streaming producers.
AI as a tool
For Mandel and Sobolewski, one promising application is making player statistics easier to access and analyze. AI tools can help identify player strengths and weaknesses, analyze gameplay footage (known as video on demand, or VODs) and scout opponents, Mandel explained.
“I think a lot of people are very doomer about AI,” Mandel said. He stressed the importance of distinguishing between generative AI and AI used particularly in data analysis and machine learning.
Mandel acknowledged concerns that AI could put people out of work, but said esports is a new industry where roles are still evolving.
“In esports, it has this very startup mentality…it’s been such a new phenomenon that the titles and jobs are being made up as we go,” he said.
Sobolewski compared AI’s potential for statistical analysis in Rocket League, a competitive game that combines soccer with vehicular acrobatics, to its use in traditional sports.
“In sports, I think everything comes down to numbers, and that’s what AI is good at, right? Every sport has something that they track, there’s statistics in everything, right? And so AI is going to be one of the things that helps with that, definitely,” said Sobolewski.
But Sobolewski and Mandel see a limit to what AI should bring to the competition.
AI as a threat
Games such as Apex Legends and Counter-Strike face significant cheating issues as more sophisticated AI-powered methods speed up cheat code development, Mandel explained.
Anti-cheat programs, which detect players modifying games, are also becoming less effective against AI tampering, he said.
“It’s much easier to poke holes and break something down than it is to build it up,” he said.
Ali Sarwary, a competitive Super Smash Bros. player and vice-president of TMU’s Smash Bros. Club, has already seen AI-assisted cheating emerge in the fighting game community.
Sarwary recalled an online Twitch stream where a streamer used an AI bot that automatically moved against opponents and selected the best option for different situations.
“It’s kind of scary,” he said, explaining that while cheating is more difficult to get away with at in-person tournaments, online players may encounter opponents they have little chance of beating.
Sobolewski said he does not tolerate cheating on his Rocket League team.
“It ends up being a very monotone game where…it removes the fun of the game…not because you lose, but because you’re not playing against…a real person,” said Sobolewski.
AI vs. human creativity
Beyond competition, some students are concerned about what generative AI could mean for creativity in esports.
Sarwary said he is disappointed by the growing use of generative AI in tournament graphics.
At TMU, the Smash Bros. Club has its own artists who create tournament graphics and loading animations, Sarwary said.
However, an artist friend of his was recently falsely accused of using AI to create tournament artwork, sparking controversy within the community.
Sarwary said the accusation was particularly upsetting after seeing how much time and effort his friend had put into the work.
“I think players and other organizers can tell when there’s like no soul in an event, or there’s no soul in [artistic] creations, because fighting games and esports in general is very, very soulful,” said Sarwary.
Mandel and Sobolewski also expressed their frustrations.
“Generative AI is an exploitation of human labour where consent was never present. These artists were never compensated for their work, and the continued use of it sets the agenda that it would be okay to allow these practices in the future,” said Mandel.
AI can process numbers and identify patterns, but it cannot necessarily replicate the creativity and human decision-making involved in competitive gaming, explained Sobolewski.
AI and the future of esports
Looking ahead, Mandel predicted a “sort of renaissance” in narrative-focused games. AI can make it easier to create visually appealing games with effective mechanics, he said, but not necessarily stories that resonate with players.
“You have to make a product that people enjoy, and if you make a product that a robot enjoys, well, I don’t think people are going to enjoy that,” he said.
Despite fears about AI’s future in gaming, Mandel encouraged students to keep an open mind.
“Don’t shy away from learning how to use [AI], because that is the only way to kind of defend yourself,” he said.
Sobolewski advised using AI as a tool but not making it your entire identity. “Just go with the flow,” he said.
For these athletes, the appeal of esports ultimately comes down to the human behind the controller.
“The beauty of e-sports…is that at any time of your life, at any moment, you can choose to become good at it. You can choose to compete. And that’s something that doesn’t exist in traditional sports,” concluded Mandel.






Leave a Reply