Skip to main content
MIT AI Risk Initiative
BackOpacity (industry opacity)
Home/Risks/G'sell (2024)/Opacity (industry opacity)
Opacity (the black box problem)
Malicious use and abuse (cybercrime)

Opacity (industry opacity)

Regulating under Uncertainty: Governance Options for Generative AI

G'sell (2024)

SourceDOI
Sub-category
Risk Domain
6Socioeconomic & Environmental
6.4Competitive dynamics

AI developers or state-like actors competing in an AI ‘race’ by rapidly developing, deploying, and applying AI systems to maximize strategic or economic advantage, increasing the risk they release unsafe and error-prone systems.

"Opacity is not solely due to the technological complexity that limits developers’ and users’ understanding of how generative models function on a technical level. It is further exacerbated by the practices of organizations and companies that are advancing the field. Many are private companies that choose to withhold from the public many of the precise characteristics of their most advanced models."(p. 69)

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

Supporting Evidence (1)

1.
"AI developers cite both near-term competition and security risks to justify withholding many vital details of their models from the general public."(p. 69)

Part of Technical and operational risks

Other risks from G'sell (2024) (33)

Technical and operational risks

7.3 Lack of capability or robustness
AI systemUnintentionalOther

Technical and operational risks > Technical vulnerabilities (Robustness - unexpected behaviour)

7.3 Lack of capability or robustness
AI systemOtherPost-deployment

Technical and operational risks > Technical vulnerabilities (Robustness - vulnerability to jailbreaking

2.2 AI system security vulnerabilities and attacks
HumanIntentionalPost-deployment

Technical and operational risks > Technical vulnerabilities (The risk of misalignment)

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

Technical and operational risks > Factually incorrect content (inaccuracies and fabricated sources)

3.1 False or misleading information
AI systemUnintentionalPost-deployment

Technical and operational risks > Opacity (the black box problem)

7.4 Lack of transparency or interpretability
OtherUnintentionalOther
View all 33 risks from this paper →