No, the AI Industry Is Not Monopolized

The reality of the AI ecosystem is perhaps more competitive and disruptive than these regulators realize, and government intervention via antitrust action could both prevent consumers from having access to beneficial innovations and further harm the small players it aims to protect.

When a new product emerges, there is often a single player who disrupts our existing paradigms of what we expect or changes the nature of the market. As a result, innovation sometimes leads to a “natural monopoly” for a first-of-its-kind company until competitors rise to consumers’ demands and catch up in the market. This is not a monopoly to worry about, and typically occurs only briefly before market forces lead competitors to respond.

We have seen this and heard these fears before when innovative technologies emerge. About two decades ago, headlines proclaimed that early internet companies were concerning monopolies that would never be caught. As ITIF’s Joe Coniglio tweeted, “Imagine saying this in 2000, when the Internet was what AI is today. You might have bet eBay Yahoo! and MySpace had it locked down.” But the reality was that competitors rapidly caught up. Ultimately, they won consumers not because the government intervened, but because of their own innovations.

Even at the top level, AI is still competitive, with new generative-AI platforms emerging. ChatGPT and Google’s Gemini both receive hundreds of millions of queries each month, while other consumers may be using slightly more specific generative-AI products — such as Dall‑E or Llama, which is now built into many of Meta’s products — to create images. As with many technologies, consumers may prefer a specific option because of its user interface or because they find the results better serve their needs.

These incredibly popular products, though, are only one type of AI product, and often rely on many other AI, machine-learning, or related products. The different elements involved in developing large language models and other AI products are not retained only by the largest players with popular consumer products but by firms of various sizes and goals. Many of these players are new and low-profile, but they may have a significant impact on the development of AI applications and the overall market.

Investors do seem to realize this is a competitive market, and many are investing in various AI startups. In February 2024 alone, AI startups received over $4.7 billion in venture-capital investments. Some of these startups may go on to be acquired, their products integrated into existing products in the tech sector or other industries, but such significant investment would be unlikely if the AI market was monopolized, with the winners already chosen. While a few companies may hog the headlines, the overall AI market is made up of companies with a range of sizes, and a significant and still growing number of competitors to the biggest players.

It is concerning to see current and former regulators rush to declare an emerging market monopolized. If we had done so in the past with the markets for other technologies at such an early phase, regulators would have likely been wrong in which companies they thought were too big to be toppled.

Unnecessary antitrust action and scrutiny could deter important investment and backfire by locking in certain first-movers more than the market would. Unfortunately, the claims that AI is or is likely to be a monopoly build on existing misguided anti-tech animus and the presumption that “big is bad” and will always win.

But as history shows us, government regulators are often poor predictors of where market disruption may be heading, and can both accidentally preserve outdated competitors or prevent even better solutions. The best takeaway for the current AI market: Early innovation spurs competition more often than it creates unassailable monopolies.