Experts call on Europe to strengthen AI regulation and become a global example

News on April 13, a group of well-known artificial intelligence experts issued a policy brief on Thursday, calling on European officials to expand the scope of artificial intelligence supervision in the "EU AI Act". More than 50 experts and institutions argue in this briefing that Europe should include general artificial intelligence (GPAI) within the scope of regulatory oversight, rather than limiting it to a narrower definition.

Signatories of the briefing include organizations such as the Mozilla Foundation, and experts such as Timnit Gebru. They believe that general artificial intelligence tools may not be designed with high-risk applications in mind, but under different circumstances, they may become more dangerous. The group specifically mentioned some generative AI tools, such as ChatGPT, which have grown in popularity over the past few months.

Regulation of the development of artificial intelligence should be considered, including how data is collected, who collects it and who trains the technology, according to Mehtab Khan, resident fellow and director of the Yale/Wikimedia Mediation and Information Initiative etc.

Khan said: "GPAI should be regulated throughout the product life cycle, not just at the application level. Simple labels of high and low risk 'inherently cannot keep up with the dynamics of technology.'"

The group recommended that European policymakers plan for future regulations, for example by avoiding limiting rules to certain types of products, such as chatbots. At the same time, they also warn that developers should not avoid liability through standard legal disclaimers.

Sarah Myers West, executive director of the AI ​​Now Institute and one of the key drivers of the policy brief, said tools like ChatGPT began to rise after an early draft of the EU AI Bill was released. "The wave of generative AI makes this bill more compelling. But even before that, there was a large class of artificial intelligence types that had no specific purpose and would also be exempted."

Myers West said: “As far as we know, the EU AI regulation will be the first comprehensive regulation on artificial intelligence. Therefore, it will set a global precedent. That is why it is particularly important, it needs to cover this type of artificial intelligence well, because it It may become a template for others to follow.” (Yi Sentence)

(This article was translated by AI and edited by NetEase for proofreading)

The above is the detailed content of Experts call on Europe to strengthen AI regulation and become a global example. For more information, please follow other related articles on the PHP Chinese website!