In the 19th century, government officials came to understand that steel was essential to both economic growth and national security, so they devised policies to preserve domestic production and prevent foreign producers from competing in the domestic market.
By the 1950s, the world had too much steel. Manufacturers replaced it with products like plastic and aluminum. Still, policymakers in Japan, India, the United States and the EU maintained domestic production capacity because steel remained essential for the construction industry and military equipment.
Even now, as governments around the world continue to invest in steel, demand continues to shrink, and the OECD predicts that overcapacity will worsen further in 2023, causing tough market conditions and exacerbating climate change.
Although steel and AI are very different, many economists see AI as a general-purpose technology that can stimulate both economic growth and innovation, and therefore policymakers need to ensure domestic capacity.
But many government officials already see AI as a critical technology essential to both national security and economic development. According to a 2023 policy and program review reported to the OECD, more than 60 countries are using taxpayer money to create, deploy or research AI. That's a lot of AI.
Policymakers in the United States, Saudi Arabia, Japan, Germany, the UK and the EU have recently announced huge public investments in AI (which follow large-scale private investments). The EU has provided $1 billion in annual funding since 2018 for AI capacity building. In March 2024, the Saudi Arabian government announced it would use about $40 billion of its $900 billion sovereign wealth fund, the Public Investment Fund, to invest in AI at home and abroad.
At a national level, these investments are understandable, but overall, they could lead to overcapacity, where the supply of AI exceeds demand.
Such excess capacity creates pitfalls in addition to the already well-known risks of AI, such as bias and inaccuracy. As countries strive to maintain their domestic AI competitiveness and market share, some may abandon excess capacity, which could make the technology more accessible to criminal organizations and malicious agents. In this case, excess capacity could lead to political instability.
Moreover, AI manufacturers require significant capital to design, develop, and deploy these systems. To attract and retain such investment, some companies and governments may choose to ignore guardrails, strategies designed to limit potential negative impacts. In this case, overcapacity may be correlated with untrustworthy or irresponsible AI.
Moreover, in competition with other countries, policymakers may hoard data or limit access to the technology, making it harder to use AI collaboratively to advance knowledge and jointly address wicked problems such as climate change. Here, excess capacity may correlate with a lack of cooperation on the use of AI.
Finally, overinvestment in AI comes with opportunity costs: without deliberate intent, such investments may come at the expense of other technologies and approaches to analyzing large amounts of data.
Excess capacity is a normal problem in national and international economies. At times, supply exceeds demand. But when various governments step in to create and maintain capacity, as they did with steel and are now doing with AI, the global ripple effects become difficult to address. Policymakers need to start addressing this potential risk in existing key international forums such as the G7, G20, and the United Nations.
See more must-read articles luck:
Opinions expressed in Fortune.com editorials are solely those of the authors and do not necessarily represent the views of luck.
