Encoded reasoning
AI systems that develop, access, or are provided with capabilities that increase their potential to cause mass harm through deception, weapons development and acquisition, persuasion and manipulation, political strategy, cyber-offense, AI development, situational awareness, and self-proliferation. These capabilities may cause mass harm due to malicious human actors, misaligned AI systems, or failure in the AI system.
"Models can employ steganography techniques to encode their intermediate rea- soning steps in ways that are not interpretable by humans [166]. Since en- coded reasoning can improve model performance, this tendency might naturally emerge and become more pronounced with more capable models."(p. 25)
Part of Model Evaluations (Interpretability/Explainability)
Other risks from Gipiškis et al. (2024) (144)
Direct Harm Domains (content safety harms)
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Violence and extremism
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Hate and toxicity
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Sexual content
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Child harm
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Self-harm
1.2 Exposure to toxic content