{"id":5640,"date":"2023-09-19T18:57:43","date_gmt":"2023-09-19T18:57:43","guid":{"rendered":"https:\/\/dailyai.com\/?p=5640"},"modified":"2024-03-28T00:42:47","modified_gmt":"2024-03-28T00:42:47","slug":"the-evolution-of-ai-regulation-where-are-we-now","status":"publish","type":"post","link":"https:\/\/dailyai.com\/fr\/2023\/09\/the-evolution-of-ai-regulation-where-are-we-now\/","title":{"rendered":"L'\u00e9volution de la r\u00e9glementation de l'IA : o\u00f9 en sommes-nous ?"},"content":{"rendered":"

In the not-so-distant past, conversations about AI were primarily confined to the realms of science fiction and academic papers.<\/strong><\/p>\n

Fast forward to today, and AI is deeply integrated into the fabric of daily life, from powering smart homes to diagnosing diseases.\u00a0<\/span><\/p>\n

But as AI technology races ahead, regulatory frameworks are playing catch-up, often lagging behind the curve of innovation.\u00a0<\/span><\/p>\n

This presents humanity with a multifaceted puzzle \u2013 how do we craft laws and regulations that are agile enough to adapt to rapid technological change yet robust enough to protect societal interests?<\/span><\/p>\n

Across the globe, regulatory efforts are far from uniform. The European Union (EU) is setting the stage with its comprehensive AI Act, which aims to categorize and control AI applications based on their risk level.\u00a0<\/span><\/p>\n

Meanwhile, the US has favored voluntary commitments over immediate legislation. <\/span><\/p>\n

China, on the other hand, wields a tight grip on AI, aligning regulation with broader political and social governance goals.\u00a0<\/span><\/p>\n

And let\u2019s not forget other jurisdictions \u2013 AI is a global phenomenon that will require near-universal legislative<\/span>\u00a0coverage.\u00a0\u00a0<\/span><\/p>\n

So, where is regulation now, and how is it expected to progress?<\/span><\/p>\n

The EU approach<\/span><\/h2>\n

The EU has adopted a stringent framework with its groundbreaking AI Act, which is expected to be fully approved by the end of the year.<\/span><\/p>\n

The AI Act specifies four tiers, including “unacceptable risk AIs,” “high-risk AIs,” “limited risks AIs,” and “low-risk AIs,” each with a specific set of guidelines and prohibitions.\u00a0<\/span><\/p>\n

Unacceptable risk AIs pose immediate or considerable dangers, such as causing physical or mental harm or infringing on human rights \u2013 and these will be outright banned.\u00a0<\/span><\/p>\n

High-risk AIs are permitted but subject to strict regulations from risk management to technical documentation and cybersecurity.\u00a0<\/span><\/p>\n

Limited and low-risk AIs, on the other hand, remain predominantly unregulated but are required to maintain transparency with users.\u00a0<\/span><\/p>\n

\"AI
EU AI Act risk levels: Source: EU.<\/figcaption><\/figure>\n

There is one reprieve in the short term, as the legislation includes a grace period of about two years for companies to align with the new regulations.<\/span><\/p>\n

The EU proposes that the original creators of AI models should be liable for how their technology is used, even when embedded into a different system by another company or developer.<\/span> There are also newly added rules for \u201cfoundation models,\u201d which include generative AI models.\u00a0\u00a0<\/span><\/p>\n

Reception to the AI Act varies \u2013 some criticize its clumsy definitions and potential to sap innovation. For instance, there is considerable uncertainty over how different AI models sort into risk categories. <\/span><\/p>\n

Additionally, smaller AI companies are worried they lack the resources to dedicate towards compliance.<\/span><\/p>\n

OpenAI CEO Sam Altman <\/span>briefly threatened<\/span><\/a> to pull his company\u2019s products from the EU market if the AI Act specified requirements that can\u2019t be realistically adhered to.\u00a0<\/span><\/p>\n

In June, 150 large companies <\/span>highlighted the Act\u2019s heavy-handed approach<\/span><\/a>, stating the current rules risk destroying European competitiveness.\u00a0<\/span><\/p>\n

Divergence from the US<\/span><\/h3>\n

A <\/span>Stanford analysis<\/span><\/a> of the AI Act discussed fundamental differences between the EU and US approaches to AI regulation. <\/span><\/p>\n

Alex Engler, a fellow in Governance Studies at The Brookings Institution, said, \u201cThere\u2019s a growing disparity between the U.S. and the EU approach [to regulating AI].”\u00a0<\/span><\/p>\n

He highlights that the EU has pushed for data privacy and transparency laws, which remain mostly lacking in the US. Disparity confuses compliance for AI companies, and the costs associated with cross-jurisdictional compliance could be overbearing.\u00a0<\/span><\/p>\n

“Corporate interests will fight tooth and nail if you have two dramatically different standards for online platforms,\u201d said Engler.<\/span><\/p>\n