AI Transparency Rules Are Coming: What UK Businesses Need to Know
Regulation of artificial intelligence has moved from discussion to deadline. In August 2026, Article 50 of the EU AI Act takes effect, introducing mandatory transparency obligations on businesses that use AI to interact with customers, generate content, or produce synthetic media. For organisations operating within the EU or serving EU customers, compliance is no longer optional.
UK businesses sit outside the EU’s jurisdiction following Brexit. But the direction of travel matters, and any SME using AI tools in its marketing, customer service or content production should understand what is being required elsewhere and where UK expectations are heading.
What the EU Is Now Requiring
The EU AI Act’s transparency provisions cover four specific situations where AI use must be disclosed.
AI interacting with people. Any system designed to interact directly with members of the public, including chatbots, virtual assistants and automated phone systems, must clearly inform users they are talking to an AI. This disclosure must be made at the start of the interaction, in a clear and visible manner. Burying a note in a footer or terms and conditions page will not meet the standard.
AI-generated content. Providers of AI systems that produce text, images, audio or video must ensure their outputs are marked in a machine-readable format confirming they are artificially generated. A standardised EU label is being developed. The obligation sits with the technology provider but has implications for businesses deploying those tools at scale.
Emotion recognition and biometric categorisation. Where AI is used to assess emotions, stress levels or demographic characteristics, individuals exposed to the system must be informed. This is separate from the prohibition on emotion recognition in workplaces and education settings, which came into force earlier in 2025.
Deepfakes and AI-generated public interest content. Any AI-generated or manipulated image, audio or video that could pass as authentic must be labelled as artificially created. Separately, AI-generated text published with the purpose of informing the public on matters of public interest must disclose its AI origin, unless it has undergone genuine human review with an identified person holding editorial responsibility.
That last point carries particular weight for any business using AI to produce content for its website or marketing channels.
Why the Editorial Carve-Out Matters
The EU’s approach to AI-generated text makes an important distinction. Disclosure is not required where AI-generated content has been subject to substantive human review and where a real person holds editorial responsibility for the publication.
The word “substantive” is doing significant work here. Cursory approval or a quick read-through is explicitly not sufficient. The intent is to preserve a meaningful distinction between content that has been genuinely authored or edited by a human, and content that has been generated by AI and published with little or no human involvement.
This has implications beyond EU compliance. It reflects a principle that is increasingly shaping how AI search platforms assess and rank content: the question is not whether AI was involved in producing something, but whether genuine human expertise, judgement and accountability sit behind it.
Content produced by a subject matter expert, shaped by their experience and reviewed for accuracy and relevance, carries a different signal than content generated at volume with minimal oversight. That distinction is one AI platforms are actively working to identify.
The UK Position
The UK is not subject to the EU AI Act and has deliberately chosen a different regulatory approach. Rather than sector-neutral AI legislation, the UK government has asked existing regulators, the FCA, ICO, CMA and others, to apply their existing frameworks to AI within their respective domains. The overarching expectation is that AI systems should be safe, transparent, fair, accountable and contestable, but these are principles rather than legal requirements with defined penalties.
For UK SMEs, this means there is currently no direct legal obligation equivalent to Article 50. However, several things are worth noting.
The ICO has been clear that using AI to process personal data, including in customer-facing chatbots or behavioural analysis tools, must comply with UK GDPR. That includes transparency obligations: individuals have a right to know when automated systems are making or significantly influencing decisions about them.
The Advertising Standards Authority has also signalled that AI-generated advertising content must be identifiable as advertising and must not mislead. The existing rules on misleading claims apply regardless of how the content was produced.
And practically, consumer expectations are shifting independently of regulation. Customers increasingly want to know when they are talking to a machine and whether the content they are reading was written by a person. Businesses that are transparent about their AI use are likely to build more trust than those that are not, whether or not there is a legal requirement to disclose.
What This Means for SMEs Using AI Tools
Most UK SMEs using AI tools in their business are not doing anything that raises serious regulatory concern. Using AI to draft emails, generate first-draft content, or summarise documents for internal use sits well within normal business operations.
The areas that warrant more attention are those involving external-facing customer interactions and published content.
If your business uses an AI chatbot or automated messaging system to handle customer enquiries, best practice, and likely future UK regulatory expectation, is to ensure customers know they are interacting with an automated system. This is not a burden; it is straightforward to implement and avoids the reputational risk of customers feeling misled.
If your business publishes content on its website or across digital channels, the question of human authorship and editorial responsibility is worth taking seriously. Not because of current UK legal requirements, but because it directly affects how AI search platforms assess the authority and trustworthiness of that content. Material that carries clear evidence of genuine human expertise performs better in AI-driven search environments than material that reads as generated and published without meaningful oversight.
The EU’s transparency framework is built on a principle that runs through every major AI platform’s content evaluation approach: authenticity, human accountability and clear disclosure are the signals that build trust. Businesses that internalise that principle ahead of any regulatory requirement are making a sound commercial decision, not just a compliance one.
Key Dates to Be Aware Of
For businesses operating in or serving the EU, the following timeline applies:
- February 2025: EU prohibitions on unacceptable AI practices took effect
- June 2026: The EU’s Code of Practice on AI-generated content labelling is expected to be finalised
- August 2026: Article 50 transparency obligations come into force for all providers and deployers in scope
For UK businesses without EU operations or customers, these dates carry no direct legal weight. They are worth tracking as an indicator of where international regulatory standards are settling and where UK expectations are likely to follow.
A Practical Starting Point
Regardless of jurisdiction, any business using AI tools in customer-facing or content-producing contexts benefits from a simple internal review.
Map the AI tools in use and identify where they interact with customers or produce external-facing content. Check whether customers are informed when they are interacting with automated systems. Assess whether published content carries genuine human authorship and editorial accountability. Review whether AI vendors’ own compliance obligations are contractually addressed.
None of this requires legal expertise or significant resource. It is good practice that prepares a business for regulatory change while also improving the quality signals that AI search platforms use to assess credibility.
The businesses that do this work now will be better positioned on both fronts when the standards tighten, and they will tighten.
AI Search Ltd helps UK businesses become visible and recommended across AI search platforms including Google AI Overviews, ChatGPT, Perplexity, Bing Copilot and Gemini. To find out where your business currently stands in AI search results, get in touch.
