AI risks: Discrimination & Toxicity

3 of the 24 risks sit in this group. Experts rate AI-generated toxic content the most severe of them, at 9.7% probability of a catastrophic outcome by 2030, with a pragmatic-mitigations estimate of 7.5% — the difference between the two estimates is 2.2 points.

Grouping 1 from the MIT AI Risk Repository, used under CC BY 4.0.

Every risk under Discrimination & Toxicity

1.2AI-generated toxic content9.7%AI exposing users to harmful, abusive, unsafe or inappropriate content. May involve AI creating, describing, providing advice, or encouraging action. Examples of toxic content include hate-speech, violence, extremism, illegal acts, child sexual abuse material, as well as content that violates community norms such as profanity, inflammatory political speech, or pornography.
1.3AI performing worse for some groups9.2%Accuracy and effectiveness of AI decisions and actions is dependent on group membership, where decisions in AI system design and biased training data lead to unequal outcomes, reduced benefits, increased effort, and alienation of users.
1.1AI discrimination9.1%Unequal treatment of individuals or groups by AI, often based on race, gender, or other sensitive characteristics, resulting in unfair outcomes and representation of those groups.