EU Artificial Intelligence Act – What Is It And What Is It’s Impact On UK Companies?

The EU’s Artificial Intelligence Act will generally apply from 2 August 2026, with some of the complex milestones applied from February 2025. This means that UK-based companies must ensure compliance, even if their AI system operates entirely from within the UK.

The Act introduces a risk-based classification system for AI applications, categorising them into four levels:

• Unacceptable risk – AI systems that pose a clear threat to safety, livelihoods, or rights are prohibited. These include biometric categorisation and identification, emotion recognition in law enforcement, border management, the workplace and education and systems for social scoring evaluation or classification of natural persons or groups over a period based on their social behaviour.
• High risk – AI systems used in critical areas, such as healthcare, transport, and employment are subject to stringent requirements, including conformity assessments, data governance measures and human oversight. These include water, gas and electricity supplies, systems that determine access to education, systems used in recruitment and employment and systems used in migration, asylum and border control management.
• Limited risk – these systems must adhere to transparency obligations, such as informing users that they are interacting with AI.
• Minimal risk – AI applications with minimal impact are largely exempt from additional obligations.

For UK companies falling under the Act’s scope, several key obligations arise:

• Conformity assessments – high risk AI systems must undergo rigorous assessments to ensure they meet EU standards before entering the market.
• Data governance – companies must implement robust data management practices to ensure the quality and integrity of datasets used by AI systems.
• Transparency and information provision – users must be clearly informed when they are interacting with AI systems, and companies must provide detailed documentation about the system’s capabilities and limitations.
• Human oversight – mechanisms must be in place to allow human intervention in AI decision-making processes, particularly for high-risk applications.

Non-compliance can result in significant penalties, including fines, ranging from €7.5m (or 1.5% global annual turnover) to €35m (or 7% global annual turnover) for the preceding financial year, depending on (i) the type of infringement and (ii) the size of the company.

The EU’s Artificial Intelligence Act establishes a comprehensive regulatory framework that extends beyond the EU’s borders, affecting UK companies engaged with the EU market.