- EU agencies are rushing through proposals this week to reach an agreement on how products like ChatGPT should be regulated.
- Germany, France and Italy oppose directly regulating generative AI models, known as “foundational models.”
The photo taken on November 23, 2023 in Frankfurt am Main in western Germany shows the logo of the ChatGPT application developed by U.S. artificial intelligence research institute OpenAI on the screen of a smartphone (left) and on the screen of a laptop computer. The words AI are shown.
Kirill Kudryavtsev | AFP | Getty Images
The European Union agreed on Friday to landmark rules on artificial intelligence. The rules are likely to be the first major regulations governing emerging technologies in Western countries.
The EU's main institutions spent a week developing proposals to reach an agreement. Issues include how to regulate the generative AI models used to create tools such as ChatGPT, and the use of biometric identification tools such as facial recognition and fingerprint scans.
Germany, France and Italy oppose direct regulation of generative AI models, known as “foundational models,” and instead support self-regulation by behind-the-scenes companies through government-implemented codes of conduct.
Their concern is that overregulation could curb Europe's ability to compete with technology leaders from China and the United States. Germany and France are home to some of Europe's most promising AI startups, including DeepL and Mistral AI.
The EU AI law is the first of its kind to specifically target AI and follows Europe's longstanding efforts to regulate this technology. The origins of this law date back to 2021, when the European Commission first proposed a common regulatory and legal framework for AI.
The law categorizes AI into risk categories ranging from “unacceptable” (meaning technologies that must be banned) to high, medium, and low risk AI.
Generative AI became a mainstream topic late last year following the general availability of OpenAI's ChatGPT. This emerged after the EU's first proposals in 2021, prompting lawmakers to reconsider their approach.
ChatGPT and other generative AI tools, such as Stable Diffusion, Google's Bard, and Anthropic's Claude, are gaining traction among AI experts and others for their ability to generate sophisticated, human-like output from simple queries using vast amounts of data. It has blindsided regulators. These have drawn criticism over concerns that they could cost jobs, create discriminatory language, and invade privacy.
clock: Generative AI helps speed up recruitment process in healthcare industry
