Generative AI tools have disrupted multiple industries and sports is no different. Use of such tools to generate real time commentary, analyze data. Sports journalism was one of the first movers in the industry to experiment with such tools to provided summaries of matches and create prompt stories. Interview responses are being processed with the help of natural language processing for editors to provide concise feedback for the same.
These tools help in even creating comprehensive published article using the stats and highlights of the game. From my personal experience I have used such tools in creating fantasy teams for competition and sports betting as well where you provide the prompt and let the generative tool predict the outcome of a game, create the best formation of a football team, or provide patterns and trends using the data. Sports betting websites and sprots streamers have also started using such these tools with advanced models and compared the outcome of their decision making with the team selected by AI. What surprised me was the accuracy with which it was able to make the best team based on the stats. For fantasy players in English Premier League, xG is one of the most important stats about any player and the tool consistently created the team using this as one of its inputs. However in the long run the team could not beast the points scored by my team as there were some of the human biases and punts involved in selecting few players which was based out of the knowledge of following the sport for a while and not just data and statistics.
While it continues to disrupt sports journalism, commentary, analytics, and data processing it certainly needs improvements to expand the scope and impact the industry further.
Standardization and lack of personal human touch inhibit the possibilities of sports commentary and journalism using gen AI. Peter Drury commenting on the comeback of Ronaldo in Old Trafford, or the miracle by AS Roma against FC Barcelona hit the right chords with sports fans across the world due to the use of language and knowledge of the game which came from decades of watching it. Such tools lack such tone which can be developed using advance NLP models, video to text processing using deep learning. Another spectrum where it can be used to impact this industry is fan engagement. Stadium guided tours, sports trivia quizzes, games and analysis can be efficiently done and provided to fans around the world. Teams can create Chatbots to give walkthroughs and tours to fans and engage them. Multi-lingual capabilities and imitating famous commentators and editors like Martin Tyler, Alan Smith, Peter Drury, Harsha Bhogle can be another experience for fans while it helps them with its data analysis and processing capabilities.
