BackUnexplainable output
Unexplainable output
Risk Domain
Challenges in understanding or explaining the decision-making processes of AI systems, which can lead to mistrust, difficulty in enforcing compliance standards or holding relevant actors accountable for harms, and the inability to identify and correct errors.
"Explanations for model output decisions might be difficult, imprecise, or not possible to obtain."
Entity— Who or what caused the harm
Intent— Whether the harm was intentional or accidental
Timing— Whether the risk is pre- or post-deployment
Supporting Evidence (1)
1.
"Foundation models are based on complex deep learning architectures, making explanations for their outputs difficult. Inaccessible training data could limit the types of explanations a model can provide. Without clear explanations for model output, it is difficult for users, model validators, and auditors to understand and trust the model. Wrong explanations might lead to over-trust."
Other risks from IBM2025 (63)
Lack of training data transparency
6.5 Governance failureHumanUnintentionalPre-deployment
Uncertain data provenance
6.5 Governance failureHumanOtherPre-deployment
Data usage restrictions
7.3 Lack of capability or robustnessHumanUnintentionalPre-deployment
Data acquisition restrictions
7.3 Lack of capability or robustnessHumanUnintentionalPre-deployment
Data transfer restrictions
7.3 Lack of capability or robustnessHumanUnintentionalPre-deployment
Personal information in data
2.1 Compromise of privacy by leaking or correctly inferring sensitive informationAI systemUnintentionalPost-deployment