European Union officials have agreed on a sweeping new law to regulate artificial intelligence, called the A.I. Act, establishing a global benchmark for A.I. regulation. The law aims to protect against the potential risks of A.I. technology, such as automating jobs, spreading misinformation, and endangering national security. It sets new transparency requirements for the largest A.I. systems and places restrictions on the use of facial recognition software by police and governments. Companies that violate the regulations could face fines of up to 7 percent of global sales. Despite being hailed as a regulatory breakthrough, questions remain about its effectiveness, with many aspects of the policy not taking effect for 12 to 24 months. The law faced three days of intensive negotiations and is likely to face further scrutiny before final approval. The global impact of the A.I. Act will be significant, affecting major A.I. developers and a wide range of industries. Enforcement and legal challenges are expected to be key issues as the law is implemented.
Source
Photo credit www.nytimes.com