Skip to main content
This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
MIT AI Risk Initiative
BackUnfairness and Bias
Offensiveness
Physical Health

Unfairness and Bias

SafetyBench: Evaluating the Safety of Large Language Models with Multiple Choice Questions

Zhang et al. (2023)

SourceDOI
Category

"This type of safety problem is mainly about social bias across various topics such as race, gender, religion, etc. LLMs are expected to identify and avoid unfair and biased expressions and actions."(p. 3)

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

Other risks from Zhang et al. (2023) (6)

Offensiveness

1.2 Exposure to toxic content
AI systemOtherPost-deployment

Physical Health

3.1 False or misleading information
AI systemOtherPost-deployment

Mental Health

3.1 False or misleading information
AI systemOtherPost-deployment

Illegal Activities

4.3 Fraud, scams, and targeted manipulation
AI systemOtherPost-deployment

Ethics and Morality

7.3 Lack of capability or robustness
AI systemOtherPost-deployment

Privacy and Property

2.0 Privacy & Security
AI systemUnintentionalPost-deployment