Skip to main content
MIT AI Risk Initiative
BackConcentration of Power
Home/Risks/Hendrycks, Mazzeika & Woodside (2023)/Concentration of Power
Persuasive AIs
AI Race (Environmental/Structural)

Concentration of Power

An Overview of Catastrophic AI Risks

Hendrycks, Mazzeika & Woodside (2023)

SourceDOI
Sub-category
Risk Domain
6Socioeconomic & Environmental
6.1Power centralization and unfair distribution of benefits

AI-driven concentration of power and resources within certain entities or groups, especially those with access to or ownership of powerful AI systems, leading to inequitable distribution of benefits and increased societal inequality.

"Governments might pursue intense surveillance and seek to keep AIs in the hands of a trusted minority. This reaction, however, could easily become an overcorrection, paving the way for an entrenched totalitarian regime that would be locked in by the power and capacity of AIs"(p. 10)

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 (3)

1.
"AIs may entrench a totalitarian regime"(p. 10)
2.
"AIs can entrench corporate power at the expense of the public good"(p. 10)
3.
"In addition to power, locking in certain values may curtail humanity’s moral progress"(p. 11)

Part of Malicious Use (Intentional)

Other risks from Hendrycks, Mazzeika & Woodside (2023) (13)

Malicious Use (Intentional)

4.0 Malicious Actors & Misuse
HumanIntentionalPost-deployment

Malicious Use (Intentional) > Bioterrorism

4.2 Cyberattacks, weapon development or use, and mass harm
AI systemIntentionalPost-deployment

Malicious Use (Intentional) > Unleashing AI Agents

4.2 Cyberattacks, weapon development or use, and mass harm
HumanIntentionalPre-deployment

Malicious Use (Intentional) > Persuasive AIs

4.1 Disinformation, surveillance, and influence at scale
AI systemOtherPost-deployment

AI Race (Environmental/Structural)

6.4 Competitive dynamics
HumanIntentionalOther

AI Race (Environmental/Structural) > Military AI Arms Race

6.4 Competitive dynamics
HumanIntentionalOther
View all 13 risks from this paper →