Skip to main content
MIT AI Risk Initiative
BackAcquisition of a goal to harm society
Home/Risks/Maas (2023)/Acquisition of a goal to harm society
Situational awareness
Acquisition of goals to seek power and c…

Acquisition of a goal to harm society

Advancing AI Governance: A Literature Review of Problems, Options, and Proposals

Maas (2023)

SourceDOI
Sub-category
Risk Domain
4Malicious Actors & Misuse
4.2Cyberattacks, weapon development or use, and mass harm

Using AI systems to develop cyber weapons (e.g., by coding cheaper, more effective malware), develop new or enhance existing weapons (e.g., Lethal Autonomous Weapons or chemical, biological, radiological, nuclear, and high-yield explosives), or use weapons to cause mass harm.

"cases of AI systems being given the outright goal of harming humanity (ChaosGPT);"(p. 29)

Entity— Who or what caused the harm

Human

Due to a decision or action made by humans

AI system

Due to a decision or action made by an AI system

Other

Due to some other reason or is ambiguous

Intent— Whether the harm was intentional or accidental

Intentional

Due to an expected outcome from pursuing a goal

Unintentional

Due to an unexpected outcome from pursuing a goal

Other

Without clearly specifying the intentionality

Timing— Whether the risk is pre- or post-deployment

Pre-deployment

Occurring before the AI is deployed

Post-deployment

Occurring after the AI model has been trained and deployed

Other

Without a clearly specified time of occurrence

Part of Dangerous capabilities in AI systems

Other risks from Maas (2023) (25)

Alignment failures in existing ML systems

7.1 AI pursuing its own goals in conflict with human goals or values
AI systemUnintentionalOther

Alignment failures in existing ML systems > Faulty reward functions in the wild

7.1 AI pursuing its own goals in conflict with human goals or values
HumanUnintentionalPost-deployment

Alignment failures in existing ML systems > Specification gaming

7.1 AI pursuing its own goals in conflict with human goals or values
AI systemIntentionalOther

Alignment failures in existing ML systems > Reward model overoptimization

7.1 AI pursuing its own goals in conflict with human goals or values
AI systemIntentionalOther

Alignment failures in existing ML systems > Instrumental convergence

7.1 AI pursuing its own goals in conflict with human goals or values
AI systemIntentionalPost-deployment

Alignment failures in existing ML systems > Goal misgeneralization

7.1 AI pursuing its own goals in conflict with human goals or values
AI systemUnintentionalOther
View all 25 risks from this paper →