The Dilemma of Uncertainty Estimation for General Purpose AI in the EU AI Act

TL;DR

This paper discusses how uncertainty estimation can help ensure compliance and trustworthiness of general-purpose AI under the EU AI Act, but highlights a computational dilemma that could increase regulatory risks.

Contribution

It proposes uncertainty estimation as a key measure for legal compliance in general-purpose AI and analyzes the computational challenges involved.

Findings

Uncertainty estimation can enhance transparency and trustworthiness.

Computational costs may exceed regulatory thresholds, creating a dilemma.

Using uncertainty estimation could classify models as systemic risks.

Abstract

The AI act is the European Union-wide regulation of AI systems. It includes specific provisions for general-purpose AI models which however need to be further interpreted in terms of technical standards and state-of-art studies to ensure practical compliance solutions. This paper examines the AI act requirements for providers and deployers of general-purpose AI and further proposes uncertainty estimation as a suitable measure for legal compliance and quality assurance in training of such models. We argue that uncertainty estimation should be a required component for deploying models in the real world, and under the EU AI Act, it could fulfill several requirements for transparency, accuracy, and trustworthiness. However, generally using uncertainty estimation methods increases the amount of computation, producing a dilemma, as computation might go over the threshold ($10^{25}$ FLOPS) to classify the model as a systemic risk system which bears more regulatory burden.