US insurance regulators are examining the credit risk behind AI data centre financing as insurers become major buyers of the private debt funding this infrastructure boom. The National Association of Insurance Commissioners (NAIC) is reviewing how state-regulated insurers are exposed to these projects, a move that could force companies to hold more capital against these investments if risk is deemed understated.
The private rating pressure point
Many early-stage data centre projects seek investment-grade treatment to attract large institutional investors before facilities are complete. In theory, these ratings reflect tenant strength, lease terms, construction history, and expected cash flows. In practice, regulators are questioning whether these packages are assessed with enough caution given the swift shifts in AI demand, energy availability, and construction costs.
Regulatory override and capital rules
The NAIC gained more authority to challenge these ratings in January 2026. If a private rating diverges sharply from the regulator's own analysis, the NAIC can override it for capital treatment purposes. This matters because lower-quality investments require insurers to hold more capital, making a bond that looks efficient under one rating much less attractive if regulators decide the risk is understated.
Capital flowing through complex channels
Big technology companies and AI infrastructure providers have raised extraordinary sums through bond markets, private credit, and special-purpose vehicles. US companies issued more than $200 billion of AI-related bonds in 2025, according to the Financial Times, while a recent Axios analysis found that Alphabet, Amazon, Meta, Microsoft, and Oracle raised $255.34 billion through equity and debt in the first half of 2026. This capital flow requires careful modeling under AI for Finance frameworks to ensure accurate yield and default projections. Insurance money has become a primary channel for this capital because insurers need long-term assets to match long-term liabilities.
However, this debt carries different risks than plain corporate bonds from major tech firms. About 20 percent of US life insurers' fixed-income portfolios are now linked to private and illiquid bonds, according to Moody's Ratings. Private credit can be legitimate, but it is harder for outsiders to price and more dependent on assumptions that may not be tested until market conditions turn, a key consideration in AI for Insurance risk assessment.
Treasury scrutiny and market implications
Treasury Secretary Scott Bessent met with state insurance commissioners in May to discuss regulatory responses to life insurers' rising exposure to private assets. The department emphasized the need for regulation that manages risk while allowing market growth. If regulators force tougher capital treatment for some data centre debt, insurers may buy less of it or demand better terms. The strongest hyperscalers will still have options, while more speculative projects relying on optimistic lease assumptions may find the next funding round difficult.
Why this matters for insurance professionals
Underwriters and portfolio managers must closely monitor how the NAIC classifies data centre debt. If regulators downgrade the risk profile of these private credit investments, insurers will need to allocate more capital against them, directly affecting portfolio yields and underwriting appetites for AI infrastructure projects.
Your membership also unlocks:
