AI Usage at a Glance
Jun 1, 2017
Customer SvcPractice documented: Riot Games deployed Blitzcrank Bot, an automated ticket-handling system for its Player Support operation, which classifies incoming support tickets and provides automated responses — particularly for account recovery requests — allowing players to opt out and request a human agent.
Practice DocumentedView practice →Mar 1, 2022
Data AnalysisPractice documented: Riot Games uses deep reinforcement learning to help game designers analyze and test game balance in Legends of Runeterra before content is released to players, giving designers a tool to predict whether new cards or mechanics will be overpowered or underpowered.
Practice DocumentedView practice →May 1, 2022
Data AnalysisNew evidence: Riot Games and deep reinforcement learning in gaming | Anyscale
Evidence AddedView practice →Nov 1, 2022
ModerationPractice documented: Riot Games announced a research partnership with Ubisoft in November 2022 called 'Zero Harm in Comms,' aimed at building a shared database of in-game chat data to train AI-based moderation tools to better detect and prevent harmful player communications across both companies' games.
Practice DocumentedView practice →Nov 1, 2022
ModerationPractice documented: Riot Games deployed an automated text evaluation system across League of Legends, VALORANT, and Wild Rift in November 2022 that uses machine learning models to assess player-reported chat messages and apply penalties such as mutes and bans for harmful, offensive, or inappropriate content.
Practice DocumentedView practice →Nov 16, 2022
ModerationNew evidence: Ubisoft and Riot Games announce the 'Zero Harm in Comms' research project to detect harmful content in game chats
Evidence AddedView practice →Sep 1, 2023
Customer SvcNew evidence: Riot Games Customer Service Story | Zendesk
Evidence AddedView practice →Jan 1, 2024
ProductivityPractice documented: Riot Games announced plans to build an internal enterprise AI platform to automate and streamline workflows across business functions including Finance, People (HR), Legal, IT, and Security, with a dedicated Principal Enterprise AI Engineer role created to lead the initiative.
Practice DocumentedView practice →Jan 1, 2024
OtherPractice documented: Riot Games uses reinforcement learning and imitation learning to train AI-controlled bot opponents that players can compete against in League of Legends and 2XKO, aiming to create more realistic and challenging practice opponents than traditional rule-based bots.
Practice DocumentedView practice →Feb 1, 2024
ModerationPractice documented: Riot Games deployed Vanguard, an anti-cheat system used in VALORANT and League of Legends, which uses machine learning to detect abnormal player behaviors such as aiming anomalies, suspicious movement patterns, and unauthorized software, and can issue automated bans.
Practice DocumentedView practice →Dec 1, 2024
ProductivityPractice documented: Riot Games announced it is researching the application of generative AI to augment game development pipelines and player experiences, with a dedicated Technology Research team that prototypes tools using large language models and content generation models for use by game teams.
Practice DocumentedView practice →Jan 1, 2025
ModerationNew evidence: Future of Anti-Cheat With Riot Games - Data + AI Summit 2025
Evidence AddedView practice →Mar 1, 2025
ModerationNew evidence: AI in Esports: How Machine Learning is Transforming Anti-Cheat Systems in Esports
Evidence AddedView practice →Nov 6, 2025
OtherNew evidence: Four things we learned at this year's AI and Games Conference, 06/11/2025
Evidence AddedView practice →Nov 6, 2025
OtherPractice documented: Riot Games announced a multi-year partnership with technology firm Globant in November 2025, which includes future explorations of AI to augment esports viewing experiences for League of Legends and VALORANT fans.
Practice DocumentedView practice →Riot Games uses deep reinforcement learning to help game designers analyze and test game balance in Legends of Runeterra before content is released to players, giving designers a tool to predict whether new cards or mechanics will be overpowered or underpowered.
Riot trained a reinforcement learning agent that plays Legends of Runeterra against itself at scale to discover which decks and cards are consistently dominant or weak, producing balance scores that became part of game designers' key performance indicators. The agent's predicted balance scores were found to match design intuition and correctly predicted which decks were strongest, with directional alignment confirmed against live player data. This was Riot's first major productionized use of deep reinforcement learning for game balance and has since been extended to additional game titles.
Riot Games deployed an automated text evaluation system across League of Legends, VALORANT, and Wild Rift in November 2022 that uses machine learning models to assess player-reported chat messages and apply penalties such as mutes and bans for harmful, offensive, or inappropriate content.
Riot Games announced a research partnership with Ubisoft in November 2022 called 'Zero Harm in Comms,' aimed at building a shared database of in-game chat data to train AI-based moderation tools to better detect and prevent harmful player communications across both companies' games.
Riot Games deployed Vanguard, an anti-cheat system used in VALORANT and League of Legends, which uses machine learning to detect abnormal player behaviors such as aiming anomalies, suspicious movement patterns, and unauthorized software, and can issue automated bans.
Riot Games announced it is researching the application of generative AI to augment game development pipelines and player experiences, with a dedicated Technology Research team that prototypes tools using large language models and content generation models for use by game teams.
Riot Games announced plans to build an internal enterprise AI platform to automate and streamline workflows across business functions including Finance, People (HR), Legal, IT, and Security, with a dedicated Principal Enterprise AI Engineer role created to lead the initiative.
Riot Games announced a multi-year partnership with technology firm Globant in November 2025, which includes future explorations of AI to augment esports viewing experiences for League of Legends and VALORANT fans.
Riot Games uses reinforcement learning and imitation learning to train AI-controlled bot opponents that players can compete against in League of Legends and 2XKO, aiming to create more realistic and challenging practice opponents than traditional rule-based bots.
Have evidence about Riot Games's AI practices? Submit a report.
Submit a report →AI Trace is free and nonprofit. Support our work
Riot Games deployed Blitzcrank Bot, an automated ticket-handling system for its Player Support operation, which classifies incoming support tickets and provides automated responses — particularly for account recovery requests — allowing players to opt out and request a human agent.
Blitzcrank Bot was built using a classification system that reads Zendesk ticket data to identify the type of support request and match it to an automated response flow. At launch, it could identify 52% of account recovery tickets with approximately 93% accuracy. The system is designed to be transparent: it identifies itself as a bot and includes a prominent option for players to escalate to a human agent. Riot's support team handles approximately 3 million annual tickets, with the bot solving routine requests in under 30 seconds.
Riot Games deployed an automated text evaluation system across League of Legends, VALORANT, and Wild Rift in November 2022 that uses machine learning models to assess player-reported chat messages and apply penalties such as mutes and bans for harmful, offensive, or inappropriate content.
The system uses multiple machine learning models to classify in-game chat text, including party chat, and can detect less obvious bullying and harmful behavior beyond simple keyword matching. Riot stated it expected more than 20 times the volume of penalties compared to prior processes as a result of this system. The system assesses player behavior reports automatically and applies penalties without waiting for human review of each individual case.
