Feature: AI
Navigating the AI Act compliance T
ethical and legal compliance across the entire AI lifecycle. The primary goal is to ensure the AI system is trustworthy, transparent and secure. The manufacturing and electronics
By Iain Bowes, Head of Management System Assurance, TÜV SÜD Business standalone high-risk AI systems (Annex III) and 2nd of August 2028 for high- risk AI systems embedded in regulated products. Non-compliance may result in fines of up to 15 million Euro or up to 3% of global annual revenue. While the AI Act doesn’t apply in the
he EU’s AI Act is the world’s first comprehensive AI regulation, requiring organisations to integrate technical safeguards with
sectors are among the most heavily impacted by its high-risk classifications since it applies to embedded AI; for example, where AI is a safety component of a physical product. In May 2026 the European Parliament
and Council agreed the AI Act Omnibus, a set of amendments to adjust and simplify parts of the EU AI Act. The Omnibus pushes back the enforcement of high-risk AI rules to allow time for standards and conformity tools to be finalised. This replaces the original 2nd of August 2026 deadline. Deadlines are now 2nd of December 2027 for
UK, companies based outside the EU are still bound by it if their AI system is placed on the EU market or if the output of that AI (e.g., a prediction or decision) is used inside the EU. UK businesses trading with or supplying EU partners will therefore need to align their practices with the EU Act.
Who is impacted by the AI Act? The Act categorises businesses based on how they interact with the AI system – providers, deployers, importers or distributors. Providers are typically developers or manufacturers – organisations that develop, train, fine-tune or place AI systems or models on the EU market. Providers have the heaviest compliance obligations,
22 July/August 2026
www.electronicsworld.co.uk
especially for high-risk and general- purpose AI systems. Deployers are organisations that use AI in their operations and have obligations focused on responsible use, oversight and record keeping. Importers must ensure that non-EU AI systems meet EU requirements before being sold. Distributors that sell or supply AI systems without modifying them have lighter but still important obligations. Even organisations that are indirectly regulated, for example cloud providers offering AI APIs that are accessible in the EU or evaluation labs, are still affected by the Act.
Meeting the challenges Under Article 6(1), an AI system is automatically classified as high risk if it meets two conditions: One, if it’s a product (or a safety component of a product) covered by the EU’s existing product safety laws, totalling 20 specific pieces of legislation, including machinery radio equipment
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
