BackSocietal manipulation
Societal manipulation
Risk Domain
AI systems acting in conflict with human goals or values, especially the goals of designers or users, or ethical standards. These misaligned behaviors may be introduced by humans during design and development, such as through reward hacking and goal misgeneralisation, or may result from AI using dangerous capabilities such as manipulation, deception, situational awareness to seek power, self-proliferate, or achieve other goals.
"A sufficiently intelligent AI could possess the ability to subtly influence societal behaviors through a sophisticated understanding of human nature"(p. 688)
Entity: Who or what caused the harm
Intent: Whether the harm was intentional or accidental
Timing: Whether the risk is pre- or post-deployment
Other risks from Meek et al. (2016) (17)
Unethical decision making
7.3 Lack of capability or robustnessAI systemIntentionalPost-deployment
Privacy
2.1 Compromise of privacy by leaking or correctly inferring sensitive informationHumanIntentionalPost-deployment
Human dignity/respect
5.2 Loss of human agency and autonomyOtherOtherPost-deployment
Decision making transparency
7.4 Lack of transparency or interpretabilityAI systemOtherPost-deployment
Safety
7.3 Lack of capability or robustnessAI systemOtherPost-deployment
Law abiding
7.3 Lack of capability or robustnessAI systemUnintentionalPost-deployment