by The European Union (EU) has reached a significant milestone in its efforts to regulate artificial intelligence (AI) models, with member states and lawmakers sealing a deal after 36 hours of negotiations. The agreement, reached in Brussels, outlines restrictions on the use of AI in Europe, while also emphasizing that the rules will not impede innovation in the sector or hinder the development of future European AI leaders.
The EU’s internal market commissioner, Thierry Breton, celebrated the achievement, declaring that the EU has become the first continent to establish clear rules for the use of AI. He also highlighted the significance of the AI Act as not just a rulebook but a platform for European startups and researchers to take the lead in the global race for trustworthy AI.
The AI Act was expedited through the EU’s legislative process this year following the widespread adoption of the chatbot ChatGPT, a popular gateway to generative AI, in late 2022. While ChatGPT’s ability to produce coherent essays and poems demonstrated AI’s rapid progress, concerns arose regarding its potential misuse.
Generative AI software, including Google’s chatbot Bard, can quickly generate text, images, and audio based on simple commands in everyday language. Other examples of generative AI include Dall-E, Midjourney, and Stable Diffusion, which have the capability to create images in a wide range of styles upon request.
Negotiators faced initial difficulties during marathon talks that started on Wednesday and lasted for 22 hours, resulting only in an agreement to reconvene the following day. Negotiations resumed on Friday at 0800 GMT.
While there was no specific deadline, senior EU figures pushed for a resolution before the end of the year. The European Commission initially proposed the AI Act in 2021 to regulate AI systems in accordance with risk assessments of the software models. The law establishes that systems with higher risk to individuals’ rights or health will bear greater obligations.
The AI Act, subject to formal approval by member states and the parliament, has garnered significant attention as a pioneering global framework for the development of trustworthy AI. European Commission President Ursula von der Leyen hailed the agreement as a commitment to safety, fundamental rights, and the interests of individuals and businesses.
The EU’s concerns over AI are shared globally, evident in US President Joe Biden’s executive order on AI safety standards issued in October. While Europe is poised to implement the first comprehensive law covering the sector, China introduced specific legislation on the regulation of generative AI in August of this year.
During negotiations, one of the main challenges revolved around the regulation of general-purpose AI systems, including ChatGPT. Some member states feared that excessive regulation could hinder the growth of European AI champions like Germany’s Aleph Alpha or France’s Mistral AI.
French digital minister Jean-Noel Barrot expressed France’s intention to carefully analyze the compromise agreed upon, ensuring the preservation of Europe’s capacity to develop its own AI technologies.
The agreement incorporates a two-tier approach, establishing transparency requirements for all general-purpose AI models and imposing stricter stipulations on more powerful models. Another contentious issue involved remote biometric surveillance, specifically facial identification through camera data in public places. Governments sought exceptions for the purposes of law enforcement and national security.
While the agreement bans real-time facial recognition, it also provides for a limited number of exemptions. Despite the milestone achievement, not all parties were satisfied with the outcome. Daniel Friedlaender, Europe chief at CCIA, a major tech lobbying group, expressed concerns over the prioritization of speed over quality, potentially leading to disastrous consequences for the European economy. CCIA Europe’s policy manager, Boniface de Champris, warned that the agreement may deter the European champions that the EU seeks to empower.
To enforce the law, the EU will establish the EU AI Office, which will have the authority to monitor and penalize violators. The office will have the ability to impose fines amounting to seven percent of a company’s turnover or 35 million euros, whichever is greater.
*Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it
Ravina Pandya, Content Writer, has a strong foothold in the market research industry. She specializes in writing well-researched articles from different industries, including food and beverages, information and technology, healthcare, chemical and materials, etc. With an MBA in E-commerce, she has an expertise in SEO-optimized content that resonates with industry professionals.
