CMA's potential competition concerns for Foundation Models/Generative AI

The CMA has published an update paper building upon its initial review of Foundation Models (FMs) (the technology underpinning generative AI tools such as ChatGPT).

The CMA is concerned that the ways in which the sector is developing risk negative market outcomes. It sets out three key risks to competition:

1. Firms that control critical inputs for developing FMs may restrict access to them to shield themselves from competition. 
2. Powerful incumbents could exploit their positions in consumer or business facing markets to distort choice in FM services and restrict competition in FM deployment. 
3. Partnerships involving key players could reinforce or extend existing positions of market power through the value chain.

The paper also sets out an updated set of AI Principles, with which the CMA “urges firms to align their business practices”.  These are:

• Access - Ongoing ready access to inputs
• Diversity - Sustained diversity of business models and model types
• Choice - Sufficient choice for businesses and consumers so they can decide how to use FMs
• Fair Dealing - No anti-competitive conduct
• Transparency - Consumers and businesses have the right information about the risks and limitations of FMs
• Accountability - FM developers and deployers are accountable for FM outputs

The CMA provides some further details about each principle (which you may be able to make out in the image taken from the report shown below), noting for example that “Open-source models can help reduce barriers to entry and expansion” and the importance that consumers and businesses “are not locked into one provider or ecosystem”.

The CMA also flags several further existing workstreams it has on AI:

• It intends shortly to publish a more detailed technical update report including detail on market developments since its initial report on FMs and feedback received from stakeholders
• It is currently examining a number of potential critical inputs to FM-related markets through its Cloud Market Investigation and its intended examination of the competitive landscape in AI accelerator chips (part of the next phase of its work on FMs)
• It plans to monitor current and emerging partnerships closely and step up its use of merger control to examine whether such arrangements fall within the current rules (see for example its investigation of Microsoft's partnership with OpenAI - more on that here)
• It is considering how it might use the new powers anticipated in the Digital Markets, Competition and Consumers Bill to enforce consumer protection law against infringing firms should unfair practices emerge in AI-powered markets.
• It is working on a joint statement with the ICO on the interaction between competition, consumer protection, and data protection in FMs.

A further update is expected in Autumn 2024.