Governments can support incentives to adopt AI by ensuring that markets remain competitive and incumbency advantage does not translate into persistent gains which discourages technology adoption by lagging firms or potential new entrants. Ensuring that open-source solutions remain a viable alternative to closed ones would also facilitate diffusion.
Policymakers can encourage competition and innovation simultaneously, by facilitating access to AI inputs (data, compute, skills), removing investment barriers for SMEs, improving market transparency and encouraging experimentation through regulatory sandboxes. Policymakers also need to monitor market developments in real time, promote system interoperability and scrutinise vertical strategies through acquisitions and partnerships that can lock in users and foreclose rivals.
While early policy action may hamper innovation, late action risks allowing concentration to become entrenched, making remedial steps difficult to implement. International cooperation and knowledge sharing among authorities, in particular competition authorities, can help build expertise and experience by enhancing transparency. Cross-border cooperation facilitates a coordinated, timely and balanced approach given the global nature of AI markets.
Policymakers should also address the risk of regulatory fragmentation by coordinating across domestic authorities, including energy and digital regulators to ensure that the competition bottlenecks in one market, upstream, such as chips or electricity, and downstream such as digital platforms, do not harm competition in other segments of the value chain.
