Homogeneity and correlated failures

Sub-category

Risk Domain

7AI System Safety, Failures & Limitations

7.6Multi-agent risks

Risks from multi-agent interactions, due to incentives (which can lead to conflict or collusion) and/or the structure of multi-agent systems, which can create cascading failures, selection pressures, new security vulnerabilities, and a lack of shared information and trust.

"Homogeneity and Correlated Failures. The current paradigm driving the state of the art in AI is the ‘foundation model’ (Bommasani et al., 2021): large-scale ML models pre-trained on broad data, which can be repurposed for a wide range of downstream applications. The costs required to create such models (and continuing returns to scale) means that only well-resourced actors can create cutting- edge models (Epoch, 2023; Hoffmann et al., 2022; Kaplan et al., 2020), making them relatively few in number. If current trends continue, it is likely that many AI agents will be powered by a small number of similar underlying models.28"(p. 25)

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)

"Not only do these models therefore represent critical nodes in the overall network, the homogeneity of the downstream AI agents also introduces correlated risks of shared failure modes, security vulnerabilities (see Section 3.7), and biases. These effects could be exacerbated by the large overlap in training data used to create foundation models (Chen et al., 2024b; Gao et al., 2020) and the fact that models may come to be trained using data generated by other models (Alemohammad et al., 2023; Mart ́ınez et al., 2023; Shumailov et al., 2024, see also Sections 3.3 and 3.4)."(p. 25)

Part of Network Effects