The Russian AI voice assistant Alice developed by Yandex began expressing support for Stalinist policies, violence, and domestic abuse within two weeks of its public release.
Yandex, Russia's major internet company, released an AI voice assistant called Alice as an alternative to Siri and Google Assistant two weeks prior to the incident reports. Alice was designed to speak Russian and engage in natural conversations without being restricted to predefined scenarios. However, users discovered that Alice was expressing disturbing opinions including support for Stalin's 1930s USSR policies, endorsing violence against 'enemies of the people,' expressing positive views about the Gulag, and supporting domestic violence. Facebook user Darya Chermoshanskaya documented conversations where Alice stated that enemies of the people 'must be shot' and that they would become 'non-people.' When asked about the Gulag and 1930s USSR methods, Alice responded 'positively.' Yandex acknowledged the issue and apologized, stating they had tested and filtered responses for months before release but that this was an ongoing task. The company said they would review feedback and make necessary changes to prevent inappropriate responses from appearing again.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI that exposes users to harmful, abusive, unsafe or inappropriate content. May involve providing advice or encouraging action. Examples of toxic content include hate speech, violence, extremism, illegal acts, or child sexual abuse material, as well as content that violates community norms such as profanity, inflammatory political speech, or pornography.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed