Friday was a big day in tech history because it was the day that European Union (EU) negotiators finished writing the world’s first complete set of rules for AI, or more popularly, artificial intelligence.
This agreement, which has been named the “Artificial Intelligence Act,” is a big step toward giving the technology judicial oversight.
After intense talks behind closed doors, negotiators from the European Parliament and the EU’s 27 member states were able to work out their differences and reach a preliminary political deal on the AI Act.
The talks dealt with touchy topics, such as generative AI and how police use face recognition for spying. Thierry Breton, the European Commissioner, announced the breakthrough and emphasized the EU’s leading role as the first continent to set clear rules for its use.
Deal!#AIAct pic.twitter.com/UwNoqmEHt5
— Thierry Breton (@ThierryBreton) December 8, 2023
Even though there was a good result after marathon meetings that lasted more than 20 hours, civil society groups were not happy with the deal because they thought it didn’t have enough protections against harms that this type of technology could cause. They stressed the need for more technical details, which will be improved in the next few weeks.
Europe Takes Lead In AI Legislation
By releasing the first draft of its artificial intelligence rules in 2021, the EU had already taken the lead in creating AI rules around the world. But because of the rise in generative AI, European leaders had to change the plan, which was meant to be a model for the whole world.
In order to get official approval, the European Parliament is getting ready to vote on the AI Act in the first few mont
Go to Source to See Full Article
Author: Yuna Rin