The Financial Times reports that Blackstone analysts are skeptical that Anthropic, even following a potential public listing, would secure an investment-grade credit rating from rating agencies. The assessment highlights a persistent gap between the towering valuations assigned to leading AI labs in private markets and the financial fundamentals that credit analysts typically scrutinize.
Anthropic, the maker of the Claude models and OpenAI's main rival, has raised tens of billions of dollars in recent years from backers including Amazon, Google and several sovereign wealth funds, pushing its valuation into the tens of billions. But like most frontier AI developers, the company burns through substantial amounts of cash to fund model training and the cloud compute capacity it needs, a spending pattern that worries analysts focused on credit risk.
An investment-grade rating is typically reserved for companies seen as capable of repaying debt with low default risk, which generally requires stable and predictable cash flows. Anthropic's profile, marked by significant operating losses and reliance on repeated capital raises, sits at odds with that standard, even as its revenue grows quickly on the back of rising enterprise adoption of its products.
The Blackstone assessment, coming from one of the largest alternative asset and private credit managers, lands at a moment when several AI labs, including Anthropic and OpenAI, are frequently floated as candidates for a future public listing. It serves as a reminder that access to traditional bond markets, and therefore to cheaper debt financing, could remain constrained for these companies until their path to sustained profitability becomes clearer.