AI update – CMA's approach to AI

AI update – CMA's approach to AI

Overview

The Competition and Markets Authority ("CMA") published its AI Strategic Update ("Strategic Update")1 on 29 April 2024, in response to the UK Government's White Paper on its approach to regulating artificial intelligence and machine learning ("AI") that was presented to Parliament on 29 March 20232, and call for action to the regulators (on 6 February 2024)3. The Strategic Update contained details of the CMA's workstreams in relation to AI and discussed the benefits of AI for businesses and consumers as well as the risks that AI may pose to consumers and to competition within UK markets. Within the Strategic Update, the CMA expressed its commitment to continuing "to research impacts on competition and consumer protection in AI-related markets"4.

Other regulators have also provided responses to the White Paper, including the Financial Conduct Authority which issued its AI Update, and the Bank of England and the Prudential Regulation Authority which issued a joint letter setting out their approach, both on 22 April 20245.

Background

In February 2024, the Government issued a response to its consultation on its AI regulation White Paper, which set out initial proposals to develop a pro-innovation regulatory framework for AI that would operate in lieu of a standalone piece of cross-sectoral AI regulation6. The framework outlined five cross-sectoral principles ("Principles") aimed at ensuring safety, security and robustness, appropriate transparency and explainability, fairness, accountability and governance, and contestability and redress. The Government also sought to establish a central function to facilitate effective risk monitoring and regulatory coordination.

The Government then issued the call for action requesting that all regulators that consider AI to be relevant to their work7 publish their strategic approach to AI regulation by 30 April 20248. Individual letters sent by the Department for Science, Innovation and Technology to a selection of leading regulators requested that these publications include details of the measures that will be taken to align each regulator's AI plans with the Principles, as well as provide an analysis of AI-related risks within their regulated sectors, an explanation of their existing capacity to manage AI-related risks and a plan of activities for the next 12 months, including the publication of addition AI guidance where necessary.

The CMA has been increasingly considering the potential impact of various aspects of AI over recent years, publishing a paper on algorithms and algorithmically-driven AI systems in 20219 and identifying AI foundation models (FMs) as an important emerging technology and possible driver of competition and consumer risks in 2022, subsequently publishing an initial report (on 18 September 2023)10 and an update paper (on 11 April 2024)11 on FMs. The CMA plans to continue its research on the impact on competition and consumer protection in AI-related markets and is currently analysing the impact of AI on choice architecture within digital markets (for example, see its discussion paper published in April 202212 and joint paper with the Information Commissioner's Office, published on 9 August 2023)13. More broadly, the Technology Horizon Scanning Function of the CMA has been analysing incoming trends within digital markets, publishing a report on 14 December 202314on ten trends in digital markets and how they may develop over the next five years and beyond. The CMA is also ensuring that it has access to the skills and experience necessary to stay up to date with the rapid pace of change in AI markets and potential AI deployment, assisted by its specialist Data Unit, nine Digital Experts, and Technology and Business Services Team, and is engaging with international networks and the Digital Research Cooperation Forum's AI and Digital Hub.

The CMA's Response

In its Strategic Update, the CMA recognised its "vital role in ensuring that consumers, businesses, and the wider economy reap the benefits of developments in AI, while harms are mitigated"15 and fully endorsed the House of Lords' recommendation (published on 2 February 2024) that ‘market competition’ should be an explicit policy objective of the Government’s work on AI16. The CMA indicated how it will continue its research into AI-related markets and will remain vigilant of new risks that arise as AI develops, ensuring that it has access to the necessary skills and experience to identify and address concerns. In particular, the CMA highlighted the incoming new pro-competition regime for digital markets and the newly appointed nine Digital Experts (acting as independent advisors to the CMA) as exciting developments that will assist the CMA in advancing its regulatory functions and its technical expertise.

Risks Identified17

Within its analysis of AI-related risks within its sector, the CMA has considered how "firms’ misuse of AI and other algorithmic systems, whether intentionally or not, can create risks to competition often by exacerbating or taking greater advantage of existing problems and weaknesses in markets". Examples are given of AI systems giving undue prominence to certain choices for customers, giving rise to a means of setting prices that could facilitate collusion and higher prices, and allowing for incumbent firms to target customers that engage with personalised offers (to the detriment of new entrants). The CMA's ultimate concern is that a "small number of the largest incumbent technology firms, with existing power in the most important digital markets, could profoundly shape the development of AI-related markets to the detriment of fair, open and effective competition."

The CMA's response also considers the risk that AI may facilitate unfair consumer practices, through exposing customers to significant levels of false and misleading information and through AI-enabled personalisation unfairly targeting vulnerable customers. Overall, the CMA acknowledges that "insufficient transparency and unclear accountability in the AI value chain may prevent deployers and consumers from understanding the risks and limitations of AI or AI-enabled services".

CMA's AI Principles18

The CMA has developed six principles and urges firms to align their business practices with them in order to shape positive market outcomes. The CMA explains that its "principles are intended to complement the UK Government’s approach and its cross-sectoral AI principles but are focused (per the CMA’s remit) on the development of well-functioning economic markets that work well from a competition and consumer protection perspective". The six principles largely reflect those previously published by the CMA in its 2023 report on AI FMs19, with the exception of the principle of flexibility regarding the use of FMs (which has been removed), and the inclusion of accountability as a standalone principle.

The CMA's six principles are:

  1. Ongoing ready access to inputs without undue restrictions;
     
  2. Sustained diversity of business models and model types (including open and closed source models);
     
  3. Sufficient choice for businesses and consumers on their use of FMs and deployment options;
     
  4. Fair dealing and tackling anti-competitive conduct so that consumers may have confidence in the products and services available and in the firms supplying these;
     
  5. Transparency for consumers and businesses as to the risks and limitations of FMs; and
     
  6. Accountability of businesses in fostering a competitive market.

Next steps for the next 12 months20

The CMA will continue with its dedicated programme of work considering the impact of FMs on markets and publish an update in Autumn 2024. This will include a forthcoming paper on AI accelerator chips, research on consumers' understanding and use of FMs (jointly with the Digital Regulation Cooperation Forum) and research into the interaction between competition, consumer protection and data protection in FMs (jointly with the Information Commissioner's Office). The CMA is also considering issuing proactive guidance to firms on how to comply with consumer law in AI-related markets and will be supporting firms to safely innovate with AI as part of the ongoing Digital Regulation Cooperation Forum's AI and Digital Hub pilot.

The CMA has identified other potential areas for further consideration including digital activities that are critical inputs for developing FMs and digital activities that are critical access points or routes to market and deploy FMs, such as mobile ecosystems and productivity software. The CMA will also be looking at its own digital transformation and ways of advancing how it works through AI.

Finally, the CMA will be progressing its ongoing Cloud Market Investigation, as well as monitoring current and emerging partnerships closely, and examining whether such arrangements give rise to competition concerns under the merger control rules (where relevant). The upcoming enactment of the Digital Markets, Competition and Consumers Bill ("DMCC Bill") will assist with this, through giving the CMA new powers directly to enforce consumer protection law against infringing firms as well as impose significant financial penalties for non-compliance. The DMCC Bill will also create a new pro-competition regime for digital markets through the introduction of 'strategic market status' in respect of a digital activity and its consequences.

Our takeaway thoughts

Throughout its response, CMA expresses its commitment to its role of identifying those firms using AI to engage in anti-competitive behaviour and unfair consumer practices, particularly through use of its merger control and digital markets regimes. The CMA will be continuing its work on understanding the impact of FMs on markets as well and we look forward to the results of its upcoming research and proactive guidance to firms on compliance with consumer law in AI-related markets.

In light of the Government's framework and the responses from the various sectorial regulators, firms should ensure that they remain aware of the regulatory developments in relation to AI. It is important for firms to understand the Government's Principles, now complemented by the CMA's six principles, and align their practices accordingly. However, although this approach may be praised for being flexible, non-prescriptive and fit for an evolving AI landscape, it may come at the cost of being unable to easily define the regulatory goal posts. However, as AI rapidly evolves worldwide, the trajectory of AI strategy in the UK may also shift. Stay tuned for updates.

Key contacts

Our experts at Stephenson Harwood are well placed to assist regulated firms understand how to meet their obligations when developing and deploying AI into their operations and supply chains.
  

CMA AI strategic update - GOV.UK (www.gov.uk) – See 'Continuing Research'

7 Such regulators include the Information Commissioner’s Office, the Competition and Markets Authority, the Equality and Human Rights Commission, the Health and Safety Executive, the Office for Product Safety and Standards, Ofcom, the Financial Conduct Authority, the Bank of England, the Medicines and Healthcare products Regulatory Agency, Ofgem, the Office for Nuclear Regulation, the Legal Services Board, Ofsted and Ofqual

 20 CMA AI strategic update - GOV.UK (www.gov.uk) at 'The CMA's next steps'