OpenAI Weighs Price Cuts as Anthropic's Claude Revenue Surge Forces Its Hand

OpenAI is weighing significant cuts to what it charges for tokens, driven in part by the competitive pressure now coming from Anthropic, according to a Wall Street Journal report published Thursday. The move would represent one of the more direct acknowledgments yet that the two companies are in a genuine pricing war, and the timing, with both firms pursuing public listings within months of each other, makes the calculus especially awkward.

Anthropic's revenue run rate crossed $47 billion in May, up from roughly $9 billion at the end of 2025. The surge has been driven largely by enterprise deals and the breakout adoption of Claude Code among software teams. That pace of growth, combined with Anthropic's recent $65 billion Series H round at a $965 billion valuation, put the company briefly above OpenAI in private-market investor rankings. OpenAI, for its part, has its own IPO in motion and is projecting profitability by 2030, though it currently burns cash at roughly 14 times the rate of its rival.

What the Price War Would Actually Mean

The pressure to cut is structural. Claude Code's adoption among enterprise developers has given Anthropic a toehold in workflows that once belonged almost entirely to OpenAI-powered tools. When engineers choose a coding assistant, they tend to standardize across their organization, which means losing a developer team to Claude is not a single lost sale but a recurring revenue deficit. OpenAI's calculation, according to the Journal, is that price cuts could stop or reverse that movement before it hardens into permanent switching behavior.

The complication is that running large AI models at scale is expensive, and cutting prices compresses margins at exactly the moment both companies need clean financial stories for their IPO filings. Anthropic has structured its compute access through a $35 billion debt facility backed by Apollo and Blackstone rather than tying up equity capital in hardware, which gives it some insulation. OpenAI's infrastructure costs are a different matter, and a price war that benefits customers could prove particularly painful for a company that does not yet break even.

The IPO Complication

Anthropic filed confidentially for a public offering on June 1, 2026, with OpenAI following days later. Both companies are racing toward listings, and the sequence matters: Anthropic's numbers are strong enough to set a high bar for what an AI company can claim as revenue, but the existence of a price war would immediately raise questions about long-term margin sustainability for any analyst reviewing a prospectus.

Anthropic's leadership has not commented publicly on whether it plans to respond with its own cuts, but it has been transparent about its pricing philosophy. At the Bloomberg Technology Conference last week, President Daniela Amodei said the company tries to plan for strong outcomes while not overextending on compute commitments. That approach has given Anthropic more pricing flexibility than rivals who have committed heavily to owned infrastructure, but it does not mean the company is immune to a race to the bottom on API costs.

"OpenAI is mulling sharp price cuts to its artificial intelligence offerings as it looks to woo consumers away from rival Anthropic." Wall Street Journal, June 11, 2026

Beyond the Pricing Surface

The deeper question behind this move is whether price alone can move enterprise customers who have already standardized on Claude. Anthropic's revenue surge through Claude Code reflects integration at the workflow level, not just API consumption. Switching an engineering organization's coding infrastructure is not simply a function of which provider charges less per token; it involves retraining, toolchain changes, and internal approvals. A price cut that shaves 20 percent off token costs may not overcome a migration cost that runs into the millions of dollars for a large enterprise.

OpenAI also competes with Anthropic across product lines beyond the API. The two companies have overlapping offerings in consumer assistants, enterprise subscriptions, and agentic platforms. If OpenAI cuts token prices, the effect could be disproportionately felt in developer and startup segments, where cost sensitivity is highest, rather than in large-enterprise deals where total cost of ownership calculations dominate.

For observers tracking the 2026 AI IPO race, the pricing dynamic adds a new variable to the analysis. Both companies need to show durable revenue growth in their prospectuses. A price war started now would show up in financial disclosures that investors will scrutinize closely. Analysts covering the sector have flagged this tension before: the competitive incentive to cut prices runs directly against the IPO incentive to demonstrate that AI revenue is durable and expanding, not price-pressured and fragile.

The practical outcome for enterprise buyers, for now, is straightforward: the two most capable general-purpose AI providers are competing aggressively for the same contracts, and that pressure will likely produce better terms for large customers over the next several months, whatever form it ultimately takes.