Big investors hunt for tomorrow's AI winners as capex angst fades

• Hyperscalers are expected to generate $340 billion more in annual operating cash flow by 2027 versus 2025, though capex will rise by approximately $534 billion in the same period
• Wellington Management ($1.3 trillion AUM) has increased positions in hyperscalers, viewing them as 'very large beneficiaries' of the AI shift, while neocloud providers (up 50-200% recently) may be vulnerable if capacity constraints ease and pricing normalizes
• Analysts warn AI monetization needs a 5x to 13x increase to justify current spending plans, and competition will likely narrow the field with fewer winners emerging than current market players

Summary: Big Investors Hunt for Tomorrow's AI Winners as Capex Angst Fades

Major asset managers are shifting their AI investment focus from Big Tech's spending concerns to identifying long-term winners, following reassuring earnings that confirmed robust AI infrastructure demand.

Key Investment Shift:

Wellington Management ($1.3 trillion AUM) and other major investors are increasing positions in hyperscalers—the largest cloud service providers including Amazon, Microsoft, and Google. These companies are now recognized as prime beneficiaries of AI's paradigm shift due to their scale and customer relationships.

Performance Dynamics:

While hyperscalers have underperformed in 2024, neocloud providers (companies renting AI computing power) have surged dramatically—some up 50-200%—capitalizing on capacity shortages and elevated spot pricing. However, analysts warn neoclouds face vulnerability when new capacity normalizes pricing.

Financial Projections:

Reuters analysis estimates hyperscalers will generate approximately $340 billion more in annual operating cash flow by 2027 versus 2025, while capex rises roughly $534 billion. Portfolio manager Richard Clode predicts profits and cash flow will outpace capex growth by late 2027-2028.

Valuation Metrics:

Hyperscaler valuations have compressed below post-pandemic peaks, with Microsoft trading at 24.6x forward earnings (highest) and Meta at 17.6x (lowest).

Market Outlook:

Capital Group ($3.6 trillion AUM) advocates holding both chipmakers and hyperscalers as part of an expanding AI ecosystem. Data centers take 12-18 months from construction to revenue generation, with recent earnings showing an inflection point. BCA Research recommends a long-hyperscalers, short-neoclouds trade.

Challenge: LGF+ZEST estimates AI monetization needs a 5-13x increase to justify current spending plans, with market consolidation expected to reduce winners significantly.