Artificial intelligence | UK Regulatory Outlook February 2024

UK updates

Government's response to AI white paper

The UK government has published its much-anticipated consultation response to its March 2023 white paper on pro-innovation approach to AI regulation.

The response reasserts that there will be no new AI legislation for the UK. Its approach to regulating AI remains softly-softly, with the emphasis on creating an innovation-friendly regulatory landscape. Existing regulators will apply their existing powers to matters involving AI falling within their jurisdictions.

The UK government has published a series of letters addressed to a number of regulators requesting them to update their strategic approach to AI and inform about the steps they are taking in line with the expectations in the AI white paper. The regulators must report by 30 April 2024.

A significant number of further government consultations and calls for evidence that businesses may wish to engage in are planned in specific areas. See more in our Insight, which also picks up on relevant comments from the House of Lords Communications and Digital Committee's report on large language models (LLMs) and generative AI, published a few days earlier.

UK AI and copyright voluntary code of practice dropped

In its consultation response on the AI white paper, the government has confirmed that it has not been possible to agree an effective voluntary code of practice to provide clarity on the relationship between intellectual property law and AI.

It has promised that it will soon set out further proposals on the way forward and flagged that it intends to explore mechanisms for providing greater transparency so that rightsholders can better understand whether content they produce is used as an input into AI models. Our Insight explores this topic in more detail.

Guidance on AI assurance

The UK Department for Science, Innovation and Technology has published guidance on AI assurance to assist organisations with understanding of the techniques they can use to ensure the safe and responsible development and deployment of AI systems.

The guidance explains the meaning of "assurance" in the context of AI: "assurance measures, evaluates and communicates the trustworthiness of AI systems." It also explains the role of AI assurance in the current UK regulatory framework in line with the AI white paper and states that it will play a crucial role in the implementation of the five core principles outlined in the white paper. The guide outlines some of the AI assurance mechanisms at introductory level.

ICO consults on generative AI and data protection

The UK Information Commissioner's Office (ICO) has announced a series of consultations on how UK data protection law should apply in relation to generative AI. It notes that it has listened to feedback that generative AI raises distinct questions to traditional models, such as:

• what is the appropriate lawful basis for training generative AI models?
• how does the purpose limitation principle play out in the context of generative AI development and deployment?
• what are the expectations around complying with the accuracy principle?
• what are the expectations in terms of complying with data subject rights?

The first consultation looks at "The lawful basis for web scraping to train generative AI models" and is open until 1 March 2024 with an online response form.

NCSC report on impact of AI on cyber security

Please see Cyber security.

EU updates

AI Act progress

On 2 February 2024, Member State representatives at the Committee of the Permanent Representatives of the Governments of the Member States to the EU (COREPER) voted to accept the final draft of the AI Act. On 13 February 2024, the EU Parliament committees accepted the near-final text. The text will now undergo the process of legal and linguistic amendments (including sorting out numbering).

These votes at technical level mean that formal adoption of the AI Act text by the Council and Parliament will be uncontroversial, with further challenges now very unlikely. The Parliament vote is likely to be on 10 or 11 April; the date of the Council vote is not yet known.

Overall it appears that the AI Act is on track to be enacted in the coming months – our prediction is that it will become law in early summer, with the first provisions coming into force (prohibitions on some categories of AI) near the end of the year.

Commission creates AI Office

The European Commission has issued its formal decision establishing the AI Office, as part of preparations for the AI Act. It will sit within the Directorate-General for Communications Networks, Content and Technology (DG CNECT), which leads on EU digital policy.

Under the AI Act, it will have a number of functions in relation to monitoring general purpose AI models for compliance and unforeseen safety risks. It will support the Commission in developing guidance, standards, codes of practice, codes of conduct, and decisions. It will also play a key role in governance across the national AI regulatory bodies appointed in Member States, as well as convening new fora (including for AI developers and the open source community) to ensure collaboration around best practice.

The decision to establish the AI Office entered into force on 21 February 2024.

Commission consults on competition in generative AI markets

The Commission has announced a consultation gathering evidence on the state of competition in generative AI (in tandem with a consultation on virtual worlds, or the metaverse).

It has invited "all interested stakeholders" to share their views by 11 March 2024. The call for contributions explains that it is undertaking "a forward-looking analysis of technology and market trends to identify potential competition issues that may arise in these fields on how competition is working".

International updates

Guidelines on responsible implementation of AI in journalism

The Council of Europe has adopted and published guidelines on the responsible implementation of AI systems at all stages of the production of news and journalism.

The guidelines aim to assist media organisations and media professionals implementing journalistic AI systems, AI technology providers and platform companies as well as member states and national regulatory authorities on how to create conditions for the responsible implementation of AI in journalism.