Credit markets are grappling with an unprecedented wave of artificial intelligence and hyperscaler bond issuance that threatens to eclipse the dot-com era's borrowing binge, according to Man Group's latest credit outlook. The investment manager projects total AI-related debt issuance could reach $465 billion in 2026, with $400 billion in investment grade and a further $65 billion in high yield and leveraged loans. That figure would represent nearly 20% of total market issuance, substantially outpacing the dot-com peak of $85 billion in 2001, which averaged 14.5% of issuance across 2000 and 2001.
The scale of the buildout is already reshaping credit market dynamics. At the start of 2015, the five major hyperscalers—Amazon, Microsoft, Meta, Alphabet and Oracle—represented less than 1% of the bond market. Today, their footprint has expanded dramatically, with AI-related capital expenditure contributing an estimated 1% to US GDP in 2026, according to economist forecasts cited in the Man Group analysis. The sheer volume of supply is creating visible stress points across the credit complex, with four of the five hyperscalers underperforming the broader market on a spread basis year to date.
Man Group warns that current spread levels do not adequately compensate investors for the risks inherent in the AI buildout, citing supply overhang and mounting concerns around competition, construction delays driven by labour shortages, public backlash and regulatory intervention. The firm notes that while many AI-related deals benefit from robust structural protections—including amortisation features, construction guarantees and hyperscaler lease backstops—the enthusiasm for the sector has encouraged what it characterises as significant overreach, particularly among lower-quality issuers.
The transmission mechanism for broader market weakness centres on what Man Group describes as the substitution effect. As investors absorb ever-larger volumes of higher-quality AI-related supply alongside Treasuries financing fiscal deficits, wider spreads across lower-quality issuers become necessary to clear the market. The firm cautions that this dynamic risks coiling a spring across the credit complex, even if supply alone is insufficient to trigger an immediate sell-off.
Execution risk looms large across the AI investment thesis. Man Group observes that capital intensity has not yet peaked, off-balance-sheet leverage is rising, and the sustainability of returns on AI investment remains unproven, particularly for neo cloud and data centre operators further down the credit quality spectrum. The firm expresses particular concern about high yield and leveraged loan issuance, where many borrowers remain firmly free-cash-flow negative despite accessing capital markets at an aggressive pace.
The software sector's expanding footprint in leveraged loan markets underscores the breadth of AI-related credit risk. According to JP Morgan data cited in the Man Group outlook, software's share of the leveraged loan market has soared in recent years, concentrating exposure in a sector now facing heightened scrutiny following a series of high-profile defaults in private credit. Man Group notes that AI and software deals have been placed under the microscope as a result of these stress events.
While the Gulf conflict has dominated headlines in recent months, Man Group argues that credit markets have told a different story, with the AI buildout emerging as the defining narrative for fixed income investors. The firm devotes a substantial portion of its second-half 2026 credit outlook to examining AI's growing footprint across both public and private credit markets, signalling the theme's centrality to portfolio construction and risk management decisions in the current environment.
The Man Group analysis draws explicit parallels to the dot-com bubble, noting that the current cycle threatens to dwarf that period in both absolute volume and market share. The comparison extends beyond issuance metrics to encompass investor behaviour, with the firm observing that enthusiasm for AI-related credits has driven deal structures and pricing that may not fully reflect underlying execution and return-on-investment uncertainties across the sector's rapidly expanding ecosystem of issuers.