Anthropic, the maker of the Claude models and a direct rival to OpenAI, is expected to receive a non-investment-grade, or 'junk,' credit rating from major rating agencies, according to a report cited by TechCrunch. The rating would come as the company prepares for a public listing, a step several leading generative AI firms are now considering to fund their mounting capital needs.
A 'junk' rating, even at the upper end of that category, signals that agencies view the issuer as carrying greater default risk than companies rated investment-grade. In practice, this would translate into higher borrowing costs should Anthropic issue debt, and a more cautious market perception of its medium-term financial footing.
The situation reflects a broader tension within the current generative AI industry: companies like Anthropic are posting rapid revenue growth, driven by enterprise and developer adoption of their models, while simultaneously racking up substantial losses tied to the enormous costs of training and running frontier models, particularly compute spending with cloud providers. Rating agencies tend to factor this profile of heavy spending and distant profitability into their risk assessments.
Anthropic's situation mirrors a pattern seen across other leading AI labs, all grappling with the same dilemma: raising massive amounts of capital to fund a costly technological race while convincing financial markets of their long-term viability. A public offering paired with a speculative-grade rating would not necessarily derail a successful fundraising effort, but it would send a signal to investors about the financial risks baked into this business model, even for companies regarded as leaders in their field.