Morgan Stanley analysts suggest summer 2026 won't be remembered for flashy new AI tech but rather for a significant pivot in AI financing. Capital markets are evolving fast to bankroll the massive infrastructure demands these artificial intelligence efforts require.

The biggest names in tech-Microsoft (MSFT), Alphabet (GOOGL), Amazon (AMZN), and Meta (META)-plan to hike their capital expenditures by 57% next year compared to 2026. According to Morgan Stanley's equity research, they're banking on these investments to deliver returns north of 25%, a bold bet reflecting confidence in AI's transformative power.

But this push isn't without friction. Revenue hasn't caught up with these spending sprees, squeezing near-term cash flows. Morgan Stanley's fretting over a wider financing gap in 2027 that will likely prompt increased AI-related credit issuance before earnings start to close the distance.

This summer, credit spreads for top-tier hyperscalers widened significantly. Higher-grade unsecured bonds stretched about 35 basis points wider, and lower-rated names saw roughly 50 basis points at their peak, though things have tightened somewhat since then. The surge of bond issuance mainly hit the high-quality unsecured categories, while asset-backed securities linked to data centers showed milder spread changes, buoyed by steady cash flows from operational assets.

The hyperscalers maintain strong credit ratings around AA, giving them considerable leeway to fund these ventures and endure borrowing cost fluctuations. Semiconductor heavyweights like Nvidia (NVDA) and Broadcom (AVGO) feature prominently here, with Morgan Stanley noting their relatively low sensitivity to modest credit price moves thanks to anticipated high returns.

On the flip side, players such as Oracle (ORCL), sitting in the mid to lower BBB credit tier, and various data center developers face tighter balance sheets and less wiggle room for higher funding costs, making their financing dynamics a bit more tenuous.

The next chapter in AI funding is expected to zoom in on compute gear - servers, chips, and energy infrastructure. Recent highlights include Nvidia's launch of a compute infrastructure financing platform and Broadcom's announcement of a hefty $35 billion chip financing deal. Morgan Stanley sees a growing trend where blue-chip issuers bolster private capital through credit backstops, guarantees, and residual value pledges, spreading risk and crowding in bigger pools of AI infrastructure investment.

This phase of AI's evolution looks less about the tech breakthroughs themselves and more about the financial engineering and credit markets adapting at breakneck speed. The stakes are high as the industry fuels its rapid growth with increasingly complex and sizable capital commitments.