EU states approve landmark AI rules

EU member states on Friday finally came to agreement and approved new laws governing the safety and use of artificial intelligence, or AI.

The European Commission proposed the rules in 2021 after scrambling to create a package of laws following last year's launch of Open AI's chatbot Chat GPT. EU countries endorsed the draft legislation in December.

Final approval by the EU's 27 member states was considered a formality after that approval, but concerns voiced by countries such as France and Germany led to protracted negotiations on the final wording of rules.

The EU aims to set global standards for AI use in industries ranging from banking, to manufacturing, medicine and travel. Rules also address the military use of AI as well as defining security parameters.

EU commissioner: 'AI Act a historical first'

Belgium, which currently holds the EU's rotating presidency, announced the approval Friday, saying, "We are very happy to announce that ambassadors just unanimously confirmed the final compromise text found on the proposal on harmonized rules on artificial intelligence."

EU Commissioner for Internal Market Thierry Breton called the proposed Artificial Intelligence Act a historical first.

"The AI Act unleashed a lot of passion … and rightly so. Today, member states endorsed the political agreement reached in December, recognizing the perfect balance found by the negotiators between innovation and safety."

The crux of the problem in finding unanimity came down to the balance between giving companies enough maneuvering room to make the development of AI products lucrative within the EU, while at the same time establishing rules for the use of a technology that is already affecting every aspect of society.

Both Germany and France — which were lobbied hard by tech companies — voiced opposition to the original rules package agreed in December, arguing that they could prohibit the creation of "EU AI champions."

Both countries dropped their opposition this week.

Boniface de Champris, a senior policy adviser for the tech lobbying group CCIA — which represents industry giants Amazon, Apple, Google and Meta — sounded the alarm after the announcement, saying, "Many of the new AI rules remain unclear and could slow down the development and roll-out of innovative AI applications in Europe."

"The act's proper implementation," said the lobbyist, "will therefore be crucial to ensuring that AI rules do not overburden companies in their quest to innovate and compete in a thriving, highly dynamic market."

The European Parliament is scheduled to vote on the rules in March or April. They could formally become law as early as May. Implementation of individual rules could start within months with most finalized over the next two years.

AI, a blessing or curse? Both?

The explosive popularity of ChatGPT and other generative AI programs served as a wake-up call for many as the technology's possibilities became clear, both in the positive and negative.

Advocates say new AI will improve production and make life easier for those employees not replaced by the new technology.

Still, AI's shocking ability to quickly and easily produce deep fake voice and video recordings, for instance, and its potential for aiding in the fire-hosing of disinformation have set off alarm bells for others.    

Europe is keen for its businesses to develop their own products for what is already becoming a gargantuan market rather than depend on innovation from the US and China, which are currently leading the way.

js/nm (AFP, Reuters)