It’s been four years since ChatGPT was launched and artificial intelligence (AI) entered the public consciousness. A while ago, we shared an overview of the state of AI regulations at the time – but over a year later, what state are they now in?
Europe
The EU remains a frontrunner in AI technology, in part thanks to its rich culture of technological advancement. The Artificial Intelligence Act came into effect in 2024, and since then, efforts have continued to reinforce AI’s use and keep its practice responsible. The Act is still being phased in even now, and this is expected to continue into 2027.
The EU’s AI Act had a risk-based structure and aligned closely with GDPR principles. High-risk AI systems (such as those involving identification and decisions regarding human rights) must undergo rigorous assessments, documentation and reporting to protect civilians and minimize the risks of information being misused by bad actors with (and by) AI.
More recently, the EU agreed to simplify AI rules, with the intent of providing innovation-friendly guidelines for the use of AI. More specifically, privileges for small and medium-sized businesses will be extended to small mid-cap companies, and the relationships between existing laws has been clarified, including how the AI Act and the EU’s various product safety laws interact. Essentially, these updates clarify the legal rights and protections of enterprises to ensure that the rules for the use of AI are as clear as possible (and it’s likely we’ll see more updates to improve them).
Alongside this agreement was the Digital Omnibus proposal last year. It will make the implication of the AI Act easier for EU businesses while maintaining European safety and fundamental rights to align more closely with GDPR, the AI Act, and the ePrivacy Directive (a legal instrument for privacy in the digital age). Citizens will also be protected by this new agreement, prohibiting AI from generating non-consensual and intimate content of real people.
Coming in December 2027 are rules governing high-risk AI systems, which include biometrics, critical infrastructure, education, employment, migration, asylum and border control.
Global Attitudes
While there may be differences between regions, AI has been governed by global laws in ways similar to how personal data has been for the past few years. To elaborate, as the world continues to use and explore the use of AI, legislators are continuously defining who is responsible for what, what disclosures are necessary, when risk assessments may be required, and how enforcement will work in practice.
As for the USA, one of the world’s leaders in AI technology, there is still not yet a federal AI statute. Instead, individual states are establishing their own standards. Many are focused on consumer and privacy protections. 2026 has seen several of these come into effect already; for example, Texas’ Responsible Artificial Intelligence Governance Act came into effect on the 1st January, which has reinforced prohibitions on identity misuse and discriminatory practices.
California has been a pioneer for AI law in the US, and continues to influence the rest of the country with its proactivity. Earlier this year (2026), California’s AI Transparency Act and the Generative AI Training Data Transparency Act both took effect, with the laws requiring disclosure of AI-generated content and other means of regulating its usage.
Summary
While AI is slowly growing out of its ‘pioneer’ era, and its possibilities (and limitations) grow clearer each year, the industry continues to thrive overall. As we push deeper into the digital age, we’ll almost certainly see global attitudes and the rules regarding AI grow increasingly similar between regions as trends are established and the best practices become clear.
Systems such as those involving finances, healthcare, and public services are generally considered ‘higher risk’. As a result, we will see frequent updates and amendments to how exactly these laws interact with each other and the ways in which AI is regulated.
Privacy and security remain the primary concern in regard to AI, and it is expected laws will continue to address them as AI gains more traction in global business and technology culture.