The EU proposes pioneering legislation to regulate AI using a four-tier, risk-linked approach.

The European Commission (EC) on Wednesday set forth new rules and actions for governing AI with the twin objectives of guaranteeing the safety and fundamental rights of people and businesses while strengthening AI uptake, investment, and innovation across the EU. (European Commission)

While the AI regulation will ensure the framework for trustworthy AI for all Europeans, the companion Coordinated plan details policy measures and investment by the Member States to enhance Europe’s standing in the global AI landscape.

The EC’s AI regulation proposals

The proposals are segregated into four risk baskets depending upon the type of AI application. Regulations apply to each depending on the extent of risk.

Unacceptable risk: The EU would completely ban these types of AI algorithms because they present a threat to the safety, livelihoods, and rights of people. (Example: Social scoring by governments)

High risk: AI applications that can be dangerous and would be subject to strict protocols before being made available to the market. (Examples: AI-enabled transport, scoring of examinations, robotic surgery, CV-sorting software, credit scoring with bias, law enforcement, and in the administration of justice)

The EC made a special reference to remote biometric identification in this risk category. “Their live use in publicly accessible spaces for law enforcement purposes is prohibited in principle,” the proposal says. However, depending upon circumstances, exceptions would be allowed such as in the case of terrorism, missing children, or for apprehending a criminal.

Limited risk: These are AI applications posing a limited risk to people, such as chatbots. It’s enough that users are advised they are dealing with AI – so they can decide whether to continue the conversation or not.

Minimal risk: These are the most common, and usually safe applications of AI (“vast majority” says the EC). Spam filters and such like. These would be unregulated.

A proposed European Artificial Intelligence Board will facilitate the implementation of the new rules, as well as develop the standards for AI.

Observance of the rules would be supervised by national competent market surveillance authorities.

Commissioner for Internal Market Thierry Breton said: “AI is a means, not an end. It has been around for decades but has reached new capacities fueled by computing power. This offers immense potential in areas as diverse as health, transport, energy, agriculture, tourism, or cybersecurity. It also presents a number of risks. Today’s proposals aim to strengthen Europe’s position as a global hub of excellence in AI from the lab to the market, ensure that AI in Europe respects our values and rules, and harness the potential of AI for industrial use.”

Related Story:   Leaked EU Draft Shows Strict Proposals Against Misuse Of AI