YouTube's AI Algorithm Promotes Banned Andrew Tate Content to Teen Users

The event involves AI systems explicitly through YouTube's recommendation algorithms that promoted banned content. The harm includes exposure of minors to harmful, misogynistic content and the platform's failure to enforce bans effectively, which indirectly leads to harm to communities and potentially violates rights. The AI system's use and malfunction (failure to filter or restrict content) directly contributed to the harm. Therefore, this qualifies as an AI Incident due to the realized harm caused by the AI system's outputs and platform policies.

The event explicitly involves an AI system—the YouTube recommendation algorithm—that actively pushed harmful content to a minor user, despite the influencer being banned. This led to exposure of toxic, misogynistic content to a vulnerable group (teenage boys), which is a form of harm to communities and potentially violates rights to safe information environments. The harm is realized, not just potential, as evidenced by reports of boys idolizing Tate and adopting derogatory views. The AI system's use and malfunction (failure to effectively filter banned content) directly contributed to this harm. Hence, the event meets the criteria for an AI Incident rather than a hazard or complementary information.

The event involves AI systems explicitly through the use of recommendation algorithms on YouTube Shorts and Instagram Reels, which autonomously selected and pushed Andrew Tate's content to a young user's feed. The content includes harmful, misogynistic messages that can cause social and psychological harm, especially to minors, fulfilling the harm criteria to communities and individuals. The AI systems' malfunction or misuse in content moderation and recommendation led to the direct dissemination of this harmful content despite bans, showing a failure in compliance and control. This direct causation of harm through AI system use fits the definition of an AI Incident rather than a hazard or complementary information.

The YouTube recommendation algorithm is an AI system that infers user preferences and generates video recommendations. The investigation shows that this AI system is actively promoting banned and controversial content to minors, which can lead to harm to communities by influencing young users with divisive and potentially harmful material. This constitutes an AI Incident because the AI system's use has directly led to harm through the dissemination of harmful content to vulnerable groups, fulfilling the harm to communities criterion. The event is not merely a potential hazard or complementary information but a realized harm caused by the AI system's outputs.