EU agrees on world’s first AI law

Social Europe: more transparent and predictable working conditions for EU workers

The flag of the European Union. Credit: symbiot/Shutterstock.com

In an unprecedented deal, European lawmakers have reached an agreement on what will be the world’s first comprehensive legal framework for regulating the use of artificial intelligence technology. However, any legislation under the AI Act, as it’s called, won’t take effect until 2025, as the European Parliament plans to vote on proposals in early 2024.

The AI Act will set a global precedent for the regulation of the increasingly advanced technology that, until now, has evaded any political attempts to reign it in. Though the law is still awaiting its final approval, lawmakers have come to an agreement, all but guaranteeing its eventual passage.

Under the new regulations, major manufacturers of AI systems, such as OpenAI – the company behind the ChatGPT chatbot – would be required to improve transparency regarding the use of AI technology, particularly in the case of generative text and image software. Chatbots like OpenAI’s ChatGPT and image generation software capable of creating “deepfakes” – a term for heavily manipulated imagery that often uses the likenesses of real people – would be required to disclose to users that the content they’re seeing is generated by AI, The New York Times reports. 

Additionally, proposed regulation would prevent governments and law enforcement from using advanced, AI-powered facial recognition software unless national security is deemed to be at risk.

AI Is Fast, Governments Are Slow

The negotiations in Brussels took three days and finished on December 7, although talks are expected to continue as lawmakers nail down the details. 

While the deal’s architects have praised it as a necessary step to ensure that the advanced, seemingly exponentially evolving technology is used for good, questions remain about its efficacy. Timing of the law’s enactment in particular has raised doubts, as it won’t take effect for at least a year, which in AI-time might as well be an eternity. 

Artificial intelligence technology, particularly generative text and image software has witnessed unforeseeable development over the past year. It was only just last year that OpenAI released ChatGPT, which has captivated users for its advanced command of language but also frightened for its tendency to “hallucinate,” or invent incorrect responses to concrete questions.

And the Rest of the World?

The EU deal comes as other governments have failed to regulate the use of AI. The UK government has avoided passing meaningful AI legislation, adopting instead a “pro-innovation” approach. 

In October of this year, US President Joe Biden issued an executive order calling for greater transparency between the government and developers of AI technology, the development of tools to ensure that systems powered by AI are safe from external attacks, and to prevent AI from being used to manufacture dangerous biological materials.

Thank you for taking the time to read this article. Do remember to come back and check The Euro Weekly News website for all your up-to-date local and international news stories and remember, you can also follow us on Facebook and Instagram.

Author badge placeholder
Written by

Cole Sinanian

Comments