The First Laws on AI: When Society Regulates the Machines We Made
A reflection on emerging legal frameworks for artificial intelligence, including the EU AI Act and new safety laws in the US, and what they reveal about human priorities.
The First Laws on AI: When Society Regulates the Machines We Made
As artificial intelligence becomes more powerful and pervasive, societies are beginning to answer a fundamental question: how should AI be governed?
In 2024, the European Union adopted the first comprehensive legal framework for artificial intelligence — the Artificial Intelligence Act (AI Act) . This regulation introduces a risk-based approach to AI systems, aimed at ensuring safety, transparency, and respect for fundamental rights across the EU.
The AI Act places concrete obligations on developers and deployers of AI systems, particularly those classified as high-risk due to their potential impact on health, safety, or civil liberties. Rather than regulating technology in the abstract, the law attempts to regulate how AI is used in real social contexts.
For readers who want to explore the legal text in more detail, the full regulation is available via the EU’s official legal database. An easier-to-navigate version, including PDF downloads, can be found at AI Act Info or ArtificialIntelligenceAct.eu .
Other regions are taking different approaches. In the United States, there is currently no single, comprehensive federal AI law. Instead, AI governance has relied on a combination of agency guidance, state-level initiatives, and executive actions. One notable example was Executive Order 14110 , issued in October 2023 to promote the safe and trustworthy development of AI, which was later rescinded in early 2025.
This fragmented approach contrasts with the EU’s centralized regulatory model, highlighting different cultural and political attitudes toward risk, innovation, and responsibility.
These emerging legal frameworks do more than regulate technology. They reveal deeply human priorities:
- protecting individual rights and dignity,
- ensuring accountability and transparency,
- balancing innovation with harm prevention,
- and deciding who gets to define acceptable use.
Seen through the lens of WearingPrompt, AI governance is not merely a bureaucratic exercise. It is a collective negotiation over how machines — trained on human language and behavior — should reflect our values, fears, and expectations back to us.
The first laws on AI are imperfect and still evolving, but they represent the earliest concrete societal responses to the feedback we are beginning to receive from our own data.
For a broader international perspective on AI governance efforts, including policy initiatives across multiple countries, see the OECD AI Policy Observatory .
