EU begin to hash out EU AI Act details starting Tuesday

By Supantha Mukherjee and Foo Yun Chee

STOCKHOLM/BRUSSELS (Reuters) – Days after EU countries and lawmakers agreed on a provisional deal on artificial intelligence rules, experts from both sides are meeting on Tuesday to hammer out details, according to three sources familiar with the matter.

Over 11 scheduled technical meetings beginning on Tuesday, government officials and aides of lawmakers will hash out specific details such as the scope of the laws and how they will work, two of the sources said.

They will clarify the legal basis on how governments can use of AI in biometric surveillance, copyright issues posed by foundation models and how to regulate major AI systems such as ChatGPT will be on the agenda.

This comes after the European Union wrangled a political agreement last Friday after some 36 hours of negotiations spread over three days, taking steps closer to becoming the first world power to enact laws governing the rapidly expanding AI industry.

The agreed text of the act would then go through the formal editing process to prepare a consolidated version.

“The Council and the European Parliament will then formally resolve and confirm the wording … After that, it will be published in the Official Journal, initiating the sunrise period,” said Alexander Duisberg, partner at law firm Ashurst.

Countries such as France and Germany have said that they will analyze the compromises reached for the AI Act. They have earlier supported views from companies like France’s Mistral and Germany’s Aleph Alpha about regulating foundation models.

The AI Act is expected to take effect in 2026.

(Reporting by Supantha Mukherjee in Stockholm and Foo Yun Chee in Brussels, Editing by Angus MacSwan)