Why AI Is Different from Previous Technology Booms

For decades, scale has been a major factor driving the performance and valuations of technology companies. As apps, websites, online stores, and social networks expanded their user base, marginal costs declined, network effects strengthened, and pricing power increased. Valuations began to reflect long-term growth potential rather than short-term profitability.

The forces that defined technology winners in the past are unlikely to dominate AI adoption, as competitive dynamics differ along six key dimensions. First, capital expenditure is no longer a minor hurdle, but a major barrier. In previous technology waves, capital requirements were largely limited to the startup phase and were relatively modest. For example, Facebook initially raised only $500,000 in seed capital.

But those early innovations were built on top of existing infrastructure such as Linux, Apache, MySQL and PHP (aka the LAMP stack), which significantly reduced the initial costs. AI, by contrast, is extremely capital intensive. Capital investment in the industry is expected to exceed $7 trillion by 2030, as companies build data centers, expand computing capacity, and invest in specialized hardware. Unlike previous technology cycles, these investment needs will not disappear as the industry matures and may even intensify.

Moreover, these costs may never come down significantly, as data center lifecycles are often measured in years, not decades. While cloud computing has also required huge investments in general-purpose servers, AI requires an entirely new infrastructure, including GPUs and Tensor Processing Units (TPUs), to handle the massive amounts of simultaneous computation involved in training and running AI models.

Such systems are costly and power-intensive. By 2027, a single large-scale AI training cycle is expected to cost more than $1 billion dollars. Only those companies that can afford the entry price will survive, giving today’s tech giants – with their huge cash flows, robust balance sheets and access to capital markets – a decisive advantage.

Second, AI’s operating cost structure undermines traditional economies of scale. In previous technology cycles, marginal costs per user fell sharply as platforms grew. Whether it was social media, software, or ride-sharing apps like Uber, costs were spread across a growing customer base, allowing platforms to maintain high margins as they expanded.

These models were also characterized by low operating costs. Once Facebook achieved sufficient scale, the marginal cost of adding users became negligible. As a result, companies paid little attention to the cost of serving each user because it rarely threatened financial viability.

AI is changing that dynamic. Controlling marginal costs is no longer optional, as large language models and other AI systems incur significant costs with every interaction that requires billions of calculations. That’s why AI companies are focused on reducing the cost per query through specialized hardware like TPUs and developing smaller, more efficient models like China’s DeepSeek.

Scale is not enough

Athird area in which AI differs from previous technological revolutions is the weakness and fragility of network effects. Older technology platforms capitalized on self-reinforcing growth. Buyers and sellers were attracted to Amazon’s marketplace precisely because activity was already concentrated there.

AI users can easily switch between models, use several simultaneously – one for text, one for images, one for coding – or even create their own. Switching costs are low and loyalty is weak, making network effects much less influential in determining long-term AI winners.

For traditional tech companies, the combination of falling marginal costs and network effects has amplified the benefits of scale, fueling the race to attract as many users as possible. This strategy made sense for companies like Facebook, which created value by monetizing consumer attention through advertising.

Artificial intelligence companies face a different cost structure. Each new version of their product requires additional capital investment. Each additional user increases costs, especially inferencing costs. While training costs can be amortized over a larger user base and some economies of scale may emerge, increased usage still increases operating costs.

Thefourth difference is the shift from market fragmentation to instant saturation. Previously, technology platforms grew within largely isolated markets, with Google dominating search and Amazon focusing on retail. By seeking separate niches, such as college students (Facebook) and professionals (LinkedIn), companies had time to mature before competition intensified.

AI, on the other hand, is a general-purpose technology that spans a variety of industries. Because users can access it instantly through apps or application programming interfaces, companies no longer have the luxury of reaching maturity before competitors emerge. This dynamic gives AI the potential to disrupt not only individual sectors, but all existing technology business models.

Fifth, political influence now matters as much as market power. Earlier waves of innovation did not require companies to engage as closely with governments and regulators as is necessary with AI. While social media has eventually come under scrutiny for its addictive effects, the perceived risks associated with today’s new technologies are deeper and in many ways existential, given AI’s potential to cause job losses, exacerbate inequality, and undermine democratic governance. As AI companies face both market forces and political pressures, companies that can shape regulation, influence public opinion, and absorb reputational risk are more likely to succeed.

Microsoft is a prime example of such a company. In an apparent attempt to gain political and social legitimacy, the company recently promised to cover the electricity costs of its data centers so that price increases would not affect consumers.

The end of winner-take-all?

Finally, AI may be less susceptible to the “winner-take-all” dynamic. Scale, near-zero marginal costs, and strong network effects have allowed companies like Facebook, Google, Amazon, and Apple to dominate social media, search, e-commerce, and smartphones, respectively. The AI sector, at least initially, is unlikely to follow this model. Instead of converging to one monopoly winner, it may support several dominant players, each controlling a different niche.

It is certainly possible for an AI company to reach a point where its technological leadership becomes self-reinforcing and virtually insurmountable. Through continuous self-improvement and overwhelmingly superior products, or even the development of artificial general intelligence, such a company can achieve sustained market power, enabling it to dominate the field.

Until then, investors should recognize that AI is following a new strategic logic. Applying outdated technology metrics to this rapidly evolving field is not only counterproductive, but also potentially costly. Investors who rely on past experience risk becoming losers in today’s AI-driven market.

Consider equity-based rewards. Historically, stock incentives have allowed technology companies to hire and retain talent, acquire intellectual property, and expand through mergers and acquisitions. But stock options can’t pay for data centers, computing power or energy infrastructure. To meet these needs, AI companies require real investment, stable cash flows, and reliable access to capital markets.

Similarly, investors once tolerated negative margins as long as user growth was steady and advertising revenues grew. But the uncertainty associated with AI and the scale of capital investment required limits their ability to assess when that investment will pay off and how AI-driven transformations will ultimately increase margins. As a result, there is an increasing focus on strong balance sheets and demonstrable financial strength.

Thus, the race for AI leadership will not be won by the companies with the most users or the fastest growth rates. Instead, the winners will be companies that can combine superior products with financial strength and political clout.

In this sense, AI is more reminiscent of the capital-intensive industries of the mid-20th century than the low-investment technology models of recent years. With rising transaction costs and easy consumer transition from one model to another, profitability will depend on retaining elastic demand while translating political capital and regulatory influence into sustainable competitive advantage.