The European Union (EU) has agreed on the world’s first laws to regulate artificial intelligence last Friday (8), after a 37-hour negotiation between the European Parliament and EU member states.
+ GTA VI: In less than 24 hours, the trailer already has more than 85 million views
+ NASA Highlights Cosmic Dance of Stars and Dust in the Carina Nebula
Thierry Breton, the European commissioner responsible for a set of laws in Europe that will also govern social networks and search engines, covering giants like X, TikTok, and Google, described the agreement as “historic.”
The agreement puts the EU ahead of the US, China, and the UK in the race to regulate artificial intelligence and protect the public from risks, including potential threats to life that many fear the rapidly developing technology brings.
The legislation still needs to go through some final steps. However, the political agreement means its key outlines have already been defined.
The foundation of the agreement is a risk-based tiered system, where the highest level of regulation applies to machines that pose the greatest risks to health, safety, and human rights.
The highest-risk category is now defined by the number of computing transactions required to train the machine, known as “floating-point operations per second” (Flops). The GPT-4 is the only model that would fall under this new definition.
The lower level of regulation still imposes significant obligations on AI services, including basic rules on disclosing data used to teach the machine to do anything from writing a newspaper article to diagnosing cancer.
