The most revealing economic story of 2025 isn't about what companies are spending money on. It's about what they're not spending money on. While technology firms throw caution to the wind with
global venture capital funding for generative AI reaching approximately $45 billion, nearly doubling from $24 billion in 2023
, the broader business world has entered what feels like an investment hibernation. This isn't just a tale of two economies—it's the birth of a fundamentally new investment logic that splits the world between those who believe in digital transformation and those who believe in waiting it out.

Think of it like a poker game where half the table is going all-in on the same hand while the other half has folded.
Nearly 33% of all global venture funding was directed to AI companies, making artificial intelligence the leading sector for investments
, while
overall business investment is predicted to rise just 4.4% in 2025, up from 2.9% in 2024
. The numbers tell the story of an economy where excitement and anxiety coexist in the same boardroom.

The AI investment frenzy has reached genuinely extraordinary proportions.
Hyperscalers spent $106 billion in capex in the third quarter this year, representing a year-over-year growth rate of 75%
, while
consensus estimates suggest AI capex will climb to 94% of operating cash flows, minus dividends and share repurchases, in 2025 and 2026, up from 76% in 2024
. These companies are essentially betting their entire free cash flow on a technology that most of them are still figuring out how to monetize.

Meanwhile, the rest of the business world is practicing what economists politely call "cautious optimism" but what looks more like sophisticated procrastination.
Just under half of business owners (49%) expect their profits to increase, down from 55% in 2024
, and
business owners not expecting any change in profits jumped from 27% to 35%, suggesting many owners are concerned about growth
. When pressed about risks,
59% cited inflation
as their greatest concern—the kind of broad, systemic worry that encourages incremental thinking over bold moves.

This creates a peculiar economic dynamic where investment confidence has become binary. Companies either believe AI will transform everything and spend accordingly, or they believe nothing fundamental has changed and maintain defensive postures. There's remarkably little middle ground.
Middle market companies have been reevaluating technology spend since interest rates began rising in 2022, with many exercising caution when spending on growth-oriented technology, such as AI, during a time of macroeconomic uncertainty
.

The psychology here resembles what happens in financial bubbles, but with an interesting twist. Usually, bubbles are characterized by broad participation—everyone thinks they can get rich quick. This time, the participation is highly concentrated among a small group of massive technology companies, while everyone else watches from the sidelines.
The AI buildout is dominated by a handful of companies whose spending is so large that it has a macro impact, with taking a view on these companies requiring assessment of whether the macro math adds up
.

What makes this particularly fascinating is that both sides might be rational. The AI investors are betting on a winner-take-all dynamic where getting there first matters more than getting there efficiently. The cautious investors are betting that most AI investments will fail to generate returns, making patience a virtue. Historical precedent supports both views simultaneously. Previous technology revolutions did create enormous value, but they also destroyed enormous amounts of capital in the process.

The uncertainty driving this split is measurably intense.
Economic policy uncertainty reached a new peak of 8.3 standard deviations above its historical mean in mid-April 2025, with trade policy uncertainty soaring over 16 standard deviations
.
The Economic Policy Uncertainty Index reached its highest level this century, surpassing peaks during the 2008 financial crisis and the COVID-19 pandemic
. When uncertainty reaches these levels, it's natural for investment strategies to polarize between extreme caution and extreme aggression.

But here's where the conventional wisdom about investment cycles gets interesting.
Economics research shows that policy uncertainty can slow economic growth, primarily by suppressing firm investment and hiring
, and
a one standard deviation increase in economic policy uncertainty reduces business investment growth by around 1 percentage point after one year, with recent increases in uncertainty potentially reducing annual investment growth by 0.4 percentage points on average across OECD countries
. The AI companies are essentially betting they can outrun this macro headwind by riding a technological tailwind.

The most intriguing aspect of this divide is what it reveals about how businesses now think about competitive advantage. Traditional business investment seeks incremental improvements in known quantities. AI investment seeks exponential improvements in unknown quantities.
By 2028, approximately 33% of enterprise software applications are expected to incorporate agentic AI, up from less than 1% in 2024, enabling automation of an estimated 15% of daily work decisions
. If this prediction holds, the companies not investing now aren't just missing an opportunity—they're potentially building obsolete capabilities.

This creates what strategists call an asymmetric risk scenario. For AI investors, the downside is wasted capital. For non-AI investors, the downside might be irrelevance. That asymmetry explains why
97% of senior business leaders whose organization is investing in AI report positive ROI from their AI investments, with 34% of companies planning to invest $10 million or more next year
, while traditional businesses remain paralyzed by uncertainty.

The paradox is that both strategies create their own justifying conditions. AI investment creates competitive pressure that makes more AI investment necessary. Traditional caution creates market stability that makes caution seem wise. But these cycles can't coexist indefinitely. Eventually, either AI delivers the promised transformation and traditional businesses scramble to catch up, or AI disappoints and prudent capital allocation wins the day.

What's happening isn't just an investment cycle. It's the emergence of two parallel business philosophies that will define the next decade. The AI believers are building for a world that doesn't exist yet. Everyone else is optimizing for a world they hope still will. Both can't be right, but both have compelling reasons to keep believing they are.

— JB