The artificial intelligence boom is real, but the money isn't flowing where you might expect. Companies are pouring hundreds of billions of dollars into AI infrastructure, yet the biggest financial winners so far are not necessarily the companies building the technology itself. According to a new white paper from investment manager L&G, investors would be wise to follow the profits rather than the hype.
The scale of AI spending
AI investment has already surpassed the peak of the dotcom-era telecom boom. Back then, companies laid fiber-optic cables and built data centers in anticipation of the internet revolution. Today, tech giants and startups are spending similar sums on graphics processing units (GPUs), cloud capacity, and energy-hungry data centers to train and run AI models.
But history offers a cautionary tale. Transformative technologies don't always deliver strong returns for investors. The railroads, the internet, and the telecom buildout all changed the world, yet many of the companies that financed them ended up losing money. The same could happen in AI, warns L&G.
Who's actually making money?
The profits from AI are unevenly distributed. Chipmakers—the companies that design and manufacture the semiconductors essential for AI—are capturing the lion's share of the financial rewards. Their products are in massive demand, and they have pricing power. In contrast, many AI developers are struggling to turn rapid revenue growth into actual profits. They spend heavily on computing power, research, and talent, often outpacing what they earn.
This dynamic has become a growing concern for investors. The recent selloff in AI stocks has brought the issue into sharper focus. If the companies spending billions on AI can't show a clear path to profitability, the market may start to question the sustainability of the entire boom.
What it means for investors
For everyday investors, the lesson is to look beyond the headlines. A company can be at the center of a transformative trend and still be a poor investment if it doesn't generate profits. The L&G paper suggests that the biggest long-term winners may be the businesses that use cheaper AI to boost their productivity—not the ones spending billions to develop the technology.
Think of it like the gold rush: the miners often went broke, but the people selling picks, shovels, and jeans made steady money. In AI, the chipmakers are the pick-and-shovel sellers. The companies that adopt AI to cut costs and improve efficiency could be the ones that see their earnings grow over time.
Investors should also be aware of the broader market context. AI spending fears have already driven Treasury yields to their biggest monthly drop since 2023, as investors worry about the cost of all this investment. That shows how intertwined AI spending is with the wider economy.
The road ahead
The AI boom is not over, but it is entering a more mature phase. Companies are becoming more cautious about how they spend on AI, and investors are demanding clearer evidence of returns. The L&G paper is a reminder that technology and profits don't always move in lockstep.
For those looking to invest, the key is to focus on companies with strong cash flows and a clear path to profitability, rather than those with the most impressive AI demos. As the paper suggests, following the profits—not the hype—is often the smarter strategy.
In the meantime, the selloff in AI stocks may offer opportunities, but it also signals that the market is becoming more discerning. The companies that can show real earnings from AI will likely be rewarded; those that can't may be left behind.
