Artificial intelligence and antitrust law: Only the beginning?

The rapid advancement of artificial intelligence (AI) continues to reshape industries and economies worldwide. Antitrust regulators are confronted with the challenge of balancing competitive landscapes while fostering innovation. Recent statements shed light on their evolving stance, as more cases highlight the practical intersection of AI and antitrust law.

The intersection of artificial intelligence and antitrust law is certainly not a new topic. Still, the recent surge of concerns from various regulators and politicians is notable. With an increasing number of cases concerning the AI industry, the issue is more and more moving away from being a topic for academic debates and seems to be becoming an enforcement priority of some of the most prominent regulators worldwide.

Policy and theory

Last week, representatives from G7 member states (Canada, France, Germany, Italy, Japan, UK and USA) and the European Commission convened once again to discuss competition law enforcement in the digital sector. According to their joint statement, protecting competition in this arena remains an expected priority, with an emphasis on closely monitoring evolving technologies like AI.

The members particularly highlighted three points as decisive to tackle concerns in that area: leveraging enhanced skills for early-warning signals, empowering regulators with legislative tools for pre-emptive action, and fostering cooperation not only domestically but also on an international scale.

Not only G7 member states saw the need to send some general warning signals to the relevant stakeholders. Prior to the G7 statement, the Portuguese regulator outlined its view of potential risks to competition in the generative AI sector in an official statement. It identified access to data, access to cloud computing or specialized hardware and access to generative AI base models as key elements which need to be protected by competition law to have a level playing field in the future.

India’s antitrust enforcer apparently plans to undertake market studies on AI to see how any concerns of competition issues need to be addressed. France’s Finance Minister Bruno Le Maire even went a step further by putting one single company in the spotlight, saying that Nvidia’s “dominance in the AI sector” was a concern.

Theory into reality

The fact that discussions around AI are not merely theoretical is evidenced by two recent examples in which authorities had to grapple with specific questions:

• Earlier this week, the German regulator issued a press release and a case summary explaining that the cooperation between Microsoft and OpenAI was currently not subject to merger control. The regulator could affirm the existence of a “concentration” under German competition law as Microsoft’s “material competitive influence” on OpenAI is sufficient under the German regime even in the absent of control. However, as OpenAI’s German turnover was below the relevant threshold, Microsoft’s involvement could have only given rise to a notification obligation based on the transaction value threshold. For that, however, OpenAI would need to have had substantial operations in Germany, which the German regulator could not establish at the time of Microsoft’s investment. Nevertheless, the regulator sent the very clear signal that OpenAI’s activities have become substantial in 2023 and any deepening of the existing relationship with Microsoft could trigger a filing obligation after all.
  
  The German regulator generally only publishes official statements on merger control cases in exceptional circumstances. So, a press release and in particular a longer case summary basically saying that a cooperation was not notifiable must be largely understood as a clear signal to the industry more generally that mergers in these areas will be closely watched.
  
• On the behavioural front, the Italian regulator just yesterday launched a fact-finding probe into pricing algorithms in the air passenger transport sector. While pricing systems through the use of algorithms and data processing software are not new in the sector, the regulator said that the degree of automation and sophistication has increased and the use of artificial intelligence and machine learning techniques could even lead to additional issues in the future.

Back to the future?

Addressing and regulating AI extends far beyond antitrust law concerns. Yet, antitrust regulators will undoubtedly influence this future. Balancing innovation and regulation demands a delicate dance – one that fosters progress while safeguarding fair competition. This necessitates a multifaceted approach involving policymakers, regulators, industry players, and stakeholders.

Photo by Alex Knight on Unsplash