On Thursday morning local time, sources revealed that after several months of negotiations, all EU parties have basically reached an agreement on comprehensive rules for artificial intelligence regulation, which is expected to push for the official entry into force of the Artificial Intelligence Act.
Once formally enacted, this bill will become the world's first comprehensive legislation in the field of artificial intelligence, and also an important milestone in global regulation of artificial intelligence.
The world's first AI bill is nearing completion
In June of this year, the European Parliament passed a draft authorization for the Artificial Intelligence Act. Subsequently, EU member states, the European Parliament, and the European Commission conducted multiple tripartite negotiations on the bill to determine the final version.
This negotiation meeting is the fifth tripartite meeting of the European Union on the Artificial Intelligence Act. Insiders said that on Wednesday afternoon local time, tripartite negotiators from the European Commission, the European Parliament, and 27 member states reached a compromise after several hours of meeting and consultation, agreeing to a series of controls on generative artificial intelligence tools such as OpenAI's ChatGPT and Google's Bard.
This progress means that the Artificial Intelligence Act is one step closer to formal legislation. Next, EU lawmakers will need to approve the final version of the bill before April next year in order for it to successfully take effect before the European Parliament elections in June next year.
Currently, the US Congress has not made meaningful progress in regulating artificial intelligence. Therefore, this EU bill will become a milestone in global AI regulation, setting the tone for countries to regulate generative artificial intelligence tools.
Seeking a balance between regulation and development
It is revealed that the discussion on this bill within the EU was very extensive, and the meeting lasted from Wednesday afternoon until late at night, highlighting the intense debate within the EU regarding the regulation of artificial intelligence.
With the continuous popularity of generative AI tools, the differences between governments and technology companies around the world are also growing. Like other governments such as the United States and the United Kingdom, the European Union has been striving to strike a balance between protecting local AI startups (such as Mistral AI in France and Aleph Alpha in Germany) and mitigating potential social risks.
It has been proven that this is also a key sticking point in EU negotiations: some countries, including France and Germany, oppose multiple regulatory regulations on AI, claiming that they will unnecessarily hinder the development of local enterprises.
However, after the EU reached a compromise on Thursday morning, officials have become increasingly confident that the final bill will be officially completed. However, the technical details of the bill still need to be finalized at a series of follow-up meetings.
EU policymakers have proposed a plan that requires developers supporting artificial intelligence models to maintain information on how their models are trained, summarize the copyrighted materials used, and label the content generated by artificial intelligence. The EU also requires that all systems with "systemic risks" must cooperate with the committee to adopt industry codes of conduct, and must monitor and report any accidents that occur in the model.