AI Disrupts Software, But “Select SaaS Companies Should Emerge Stronger”
Even before the onset of the Iran war, the centre of gravity in capital markets had begun shifting from soft assets to hard tech and hard power. Hyperscalers face an existential crisis and are doubling down on their AI infrastructure buildout, with 2026 capex set to equal roughly 2% of U.S. GDP. Defense spending is accelerating globally, and a bullish commodity cycle appears intact.
Software companies, by contrast, have endured a bruising sell-off as investors reassess their terminal value amid AI gains.
This repricing marks the start of a more discerning phase in markets. The prospect of falling intelligence costs and stronger, productivity-led growth means value could accrue differently than in recent years—likely reshaping market leadership.
From hype to balance sheets
The AI cycle has gone from hype to real-world gains to genuine disruption. The next phase brings greater scrutiny of balance sheets and business models.
Tech giants that have long dominated equity markets through highly profitable, asset-light growth engines such as code, platforms and intellectual property are going all-in on physical, capital-intensive investments. Capacity, not demand, has become the binding constraint, with competitive necessity forcing aggressive spending on power, chips, data centers and networking.
Hyperscalers, among the most cash-generative companies in history, have largely funded these investments from operating cash flow. They can spend much more before crossing into net debt. However, operating margins are eroding quickly, and the growing use of debt issuance and off-balance-sheet financing by certain players bears watching.
AI chip spending alone could reach $1 trillion by 2030, and the most compelling opportunities lie where the capex dollars flow. Semiconductors remain the cleanest beneficiaries alongside bottlenecks in memory, power, optics, electricals and other critical AI segments.
SaaS repricing
As economic rents shift to hardware and infrastructure, software is undergoing a structural reset. AI progress lowers the cost of building software, intensifies competition and poses terminal value risk. Recent price action has been indiscriminate, but select software-as-a-service (SaaS) companies should emerge stronger. Those that control critical datasets, act as enabling platforms for AI deployment, or sit deeply within enterprise workflows maintain favourable outlooks. Put simply, mission-critical systems are here to stay.
A growing opportunity set
Globally, economic nationalism, policy recalibration and fiscal shifts are bolstering investment across defence, energy transition, and industrial capacity—all of which intersect with AI’s physical footprint. Capital is being redirected and reshaping sectors.
Investors quickly priced in Germany’s “fiscal bazooka” in 2025. However, they may have underestimated the wider geopolitical shifts taking place, highlighted in Canadian Prime Minister Mark Carney’s recent address at the World Economic Forum.
Growing urgency among the so-called “middle powers” for strategic autonomy over their defence capabilities and critical industries suggests further upside for fiscal spending and economic growth.
The war in Iran could spark further fiscal actions with countries spending even more on energy and defense.
European defence stocks, capital goods firms, semiconductor equipment manufacturers and select financials should benefit from these tailwinds.
In Asia, structural reforms and deep semiconductor ecosystems remain central to the AI value chain. Japan’s strategic industrial push and its key role in semiconductor supply chains—a position shared with Taiwan and South Korea—underpin a bullish equity outlook outside the U.S.
These shifts are broadening market leadership. Since late last year, global small-caps, global ex-U.S. value, and emerging markets equities have all benefited as investors have rotated beyond U.S. mega-caps.
Real assets become the bottleneck
Amid the ongoing macro and geopolitical recalibration, a continued scramble for perceived safe havens seems likely. The acceleration of fiscal spending across developed markets—coupled with Japanese monetary policy that remains behind the curve—dulls the appeal of long-dated sovereign bonds and strengthens the case for hard assets over fiat currencies.
Industrial metals have experienced some speculative excesses recently, but the long-term outlook remains bullish. A weakened U.S. dollar amid ballooning deficits, the threat of de facto yield curve management, and a more protectionist tenor in Washington lends support to a constructive commodity outlook.
Additionally, as the AI buildout and decline in the cost of compute set the stage for structurally higher growth, real assets could become bottlenecks. As suppliers to the world, emerging markets—especially Latin American countries and South Africa—stand to benefit.
A maturing cycle, not peak
The macro environment has become more complex—but also more investable for those seeking diversification and fundamentally driven opportunities.
The open question is no longer whether AI will radically alter the economy. It’s where and to what extent value will accrue: upstream in physical infrastructure and commodities, within enabling platforms, or in entirely new applications.
As value shifts from the intangible to the tangible, investors who remain selective, diversified and valuation-aware may see the next era as less of a threat than an opportunity.
About the author: Ritu Vohora, CFA, is Investment Specialist, Capital Markets at T. Rowe Price.
