Europe braces itself for AI Act effects

The year ahead will see big changes to how artificial intelligence is governed in Europe with the majority of rules under the European Union’s AI Act slated to take effect. 

The EU AI Act is the world’s first comprehensive law for Artificial Intelligence, establishing a risk-based framework that bans unacceptable AI risks (like social scoring), heavily regulates high-risk AI (like in hiring or critical infrastructure), and requires transparency for others (like chatbots), aiming to foster trustworthy AI while promoting innovation, with phased implementation starting in 2025 and most rules applying by 2026. It categorizes AI into unacceptable, high, limited, and minimal risk, with different compliance obligations for each, setting a global standard for safe AI development and use. 

The landmark piece of legislation entered into force in 2024 but on a gradual basis. In 2026 many of its rules are set to kick in, including its enforcement at both an EU and national level, with the bloc seeking to set a precedent for this rapidly evolving sector.

“The European Union emphasizes the trustworthy development of AI which could also be like a game changer,” Toon Calders, Professor of Computer Science at the University of Antwerp, told CGTN.

“People will trust AI if it comes from Europe more than AI from other sources because it’s like a quality seal, they know if it comes from the EU it has to follow certain regulations.”

Critics call the AI Act a straitjacket for innovation as Europe seeks to compete with the likes of China and the United States. 

To boost competitiveness and help support businesses, the European Commission recently launched its digital omnibus in an effort to trim down the bloc’s digital regulations. Under the plan some parts of the AI Act could be delayed until 2027, including rules relating to high-risk AI applications such as its use assessing loan applications or resumes.

The omnibus aims to streamline and simplify AI regulation and the European Commission says this flexibility is needed to help adapt to a digital age.

“Europe has not so far reaped the full benefits of the digital revolution and we cannot afford to pay the price for failing to keep up with demands of a changing world,” said Valdis Dombrovskis, European Commissioner for Economy and Productivity; Implementation and Simplification, at a press conference announcing the digital omnibus.

The EU says more time is needed to help member states and companies adapt to the rules of the AI Act but some experts warn there must be clarity on issues like liability when it fully takes effect.

“The main priority I think should also be enforcement because we can have wonderful legislation but if it’s not properly enforced then it may remain without any effects,” argued Jan De Bruyne, Professor of IT Law at KU Leuven. 

European officials say AI technology is vital for the bloc’s resilience, competitiveness and security and as the battle for AI dominance continues the EU is seeking to take steps to establish the rules of the game.

article source: Europe braces itself for AI Act effects – CGTN