Which Countries are Regulating AI?
Artificial intelligence (AI) has caught the attention of regulators across the Atlantic in the last five years. In the U.S., growing policy interest in AI is evident at both the federal and state levels, with 88 AI-related bills proposed at the federal level in 2022 and 134 in 2021, compared to a single proposed bill in 2016. Although the number of AI-related bills proposed in 2022 decreased from 2021, there was a sharp increase in bills passed into law from 3 to 9 bills.
At the federal level, the 116th and 117th Congress have passed three important bills, the National AI Initiative Act of 2020 (NAIIA), enacted as part of the National Defense Authorization Act for fiscal year 2021, the AI in Government Act and the AI Training Act. The purpose of the NAIIA is to ensure continued U.S. leadership in AI research and development, use of trustworthy AI systems, and preparing the U.S. workforce for the integration of AI across all sectors. The AI Government and AI Training Act focus on the use of AI within the federal government and calls on the Office of Management and Budget to provide guidance for agency use of AI. These congressional bills have also been accompanied by executive orders from the Trump and Biden administrations that build upon these initiatives.
At the state level, Maryland, California, Massachusetts and Washington stand out among the U.S. states that are making significant progress on AI regulation with each passing 5-7 AI-related bills between 2016 and 2022. Coming from both Democratic and Republican lawmakers, AI regulation in the U.S. seeks to establish stronger protections for consumers and mitigate the effects of algorithmic bias on workers.
In Europe, the European Commission, under Ursula von der Leyen’s leadership, has shown a strong commitment to the regulation of the digital economy with the Digital Services Act and Digital Markets Act, quickly followed by the AI Act in 2021. The AI Act still has to go through a phase of negotiations between the European Parliament, the Council, and the Commission to be adopted, but before negotiations on the final form of the law can begin, the current draft has to be endorsed by the whole Parliament by June 14.
At the member state level, several governments are already thinking about how they will regulate AI within their borders. Between 2016 and 2022, 57 AI-related bills were passed into law across the EU with the southern European countries taking the lead. The Portuguese, Spanish and Italian governments have passed 22 out of the 57 bills, and along with 23 of their EU counterparts, have rolled out their own National Strategy for AI. AI regulation in the EU follows a human-centered approach, which prioritizes citizens' fundamental rights, ethical frameworks and algorithmic transparency, while still promoting AI innovation and using AI to boost the economy by integrating the technology into value chains.