Riot Games to Use AI for Valorant Voice Chat Moderation, Raising Privacy Concerns

Thumbnail Image

The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.

Riot Games will begin recording and analyzing Valorant players' voice chats in North America starting July 13 to train AI language models for detecting toxic behavior. While no harm has yet occurred, the initiative raises concerns about potential privacy violations and wrongful disciplinary actions as the system develops.[AI generated]

Why's our monitor labelling this an incident or hazard?

The event involves the use of an AI system (voice evaluation system using language models) in the development and testing phase. Although no harm has yet occurred, the system's intended use to monitor and evaluate player voice chat for abusive language could plausibly lead to violations of rights (privacy, freedom of expression) or harm to community environment if errors or misuse occur. Since the system is not yet actively enforcing or causing harm but is being tested to prepare for future deployment, this constitutes an AI Hazard rather than an AI Incident. The article does not describe realized harm but highlights potential risks inherent in the AI system's future use.[AI generated]
AI principles
Privacy & data governanceRespect of human rightsTransparency & explainabilityAccountabilityFairnessRobustness & digital securitySafety

Industries
Media, social platforms, and marketingArts, entertainment, and recreationConsumer services

Affected stakeholders
Consumers

Harm types
Human or fundamental rightsReputationalPsychological

Business function:
Monitoring and quality controlCompliance and justice

AI system task:
Event/anomaly detection


Articles about this incident or hazard