EU ambassadors eventually approved the draft on Friday during a conference in Brussels.
Following a series of difficult last-minute talks, 27 EU member states have enacted historic regulations aimed at curbing the reach of artificial intelligence (AI). The legislation is described as a first for the globe.
Brussels had first put up a proposal for an AI law in 2021, but when chatbot ChatGPT gained widespread attention, the city spent most of last year racing to approve the proposal. A draft of the same had been agreed upon by EU governments and parliament back in December, so approving them after that should have been a formality.
However, a few countries, like Germany and France, voiced their worries that the AI law wouldn’t pass in the end, which prompted fresh negotiations to finalize the provisions.
After easing German and French concerns, EU ambassadors eventually approved the draft on Friday during a conference in Brussels.
AI Rules Agreed After Debate
Belgium, which is now holding the rotating presidency of the EU, said, “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.”
Thierry Breton, the head of the EU’s information policy, praised the “historical, world-first, pioneering” regulations.
To be fair, the AI Act aroused a great deal of passion. Recognizing the ideal balance that the negotiators struck between innovation and safety, today’s member states backed the political agreement agreed in December, the speaker added.
Berlin and Paris have shown a strong desire to safeguard AI companies in order to avoid future “European AI champions” being hampered by regulations.
Late in January, the EU’s executive branch introduced a set of pro-innovation initiatives aimed at helping European startups develop AI in an effort to alleviate their worries.
With the use of straightforward commands in common English, ChatGPT has showcased the astounding advancements in generative AI technology, which can instantly produce text, graphics, and sounds.
OpenAI, a U.S.-based company, has developed an accessible chatbot that can pass legal and medical exams and compose elegant essays and stirring poems in a matter of seconds.
The threats AI poses have become rapidly apparent to the public, ranging from the spread of modified films and photos to the accelerated spread of misinformation online.
However, it’s also become evident that Europe wants to take advantage of the enormous new prospects that generative AI, which also includes Google‘s chatbot Bard, opens.
Germany Hails the Agreement
Germany has agreed to the deal. Germany’s Economy Minister Robert Habeck stated that when putting the bill into effect, Berlin would prioritize “ease of innovation, legal clarity for businesses and necessary light and unbureaucratic structures.”
Volker Wissing, Germany’s Minister of Digital Affairs, praised himself earlier this week for “improving” the text “for small and medium-sized enterprises” in a way that would “avoid disproportionate requirements and ensure that we remain internationally competitive.”
AI Law Heads to Parliament Voting Before May Approval
The European Parliament is scheduled to vote on the draft in March or April before it finally becomes law. It should then receive formal approval in May. This law will not be effective immediately. Some of the rules will go into effect within six months, while others will take effect two years after that.
One of the major tech lobbying groups, CCIA, expressed caution but acknowledged that Friday’s approval was an “important step” towards the text’s official adoption.
According to Boniface de Champris, Senior Policy Manager at CCIA Europe, “Many of the new AI rules remain unclear and could slow down the development and roll-out of innovative AI applications in Europe, despite efforts to improve the final text after ‘victory’ was prematurely declared back in December.”