OpenAI is secretly plotting a significant reduction in Token fees, directly targeting its competitor Anthropic, with a price war brewing over corporate customers. Both companies are deeply mired in billions of dollars in losses, and price cuts may further erode their already precarious profit margins. More troubling is that both are on the eve of their IPOs, and how to capture market share without exacerbating losses will be the ultimate test for investor valuations.

Written by: Zhao Ying

Source: Wall Street Insights

OpenAI is brewing a substantial price reduction for its AI services, aimed directly at competitor Anthropic, forming a price war centered around corporate clients.

According to a report by The Wall Street Journal on Wednesday, insiders revealed that OpenAI is evaluating plans to significantly cut Token fees. Tokens are the basic units through which AI companies measure and charge, and this move is intended to capture market share from Anthropic while also preemptively responding to Anthropic's expected similar price cuts.

Discussions are still in flux. OpenAI CEO Sam Altman recently admitted at an event that the cost of AI usage has become "a huge problem" and stated, "We will have many ways to help users get more value for less spending."

This price cut discussion poses financial pressure on both companies. OpenAI and Anthropic are currently losing billions due to the enormous computing costs required for AI systems to handle queries and perform tasks, and a significant price cut may further erode their already pressured profit margins.

Anthropic's Rise, OpenAI Under Pressure

OpenAI's price cut considerations reflect the competitive pressure it faces in the battle for corporate clients.

Anthropic's programming tool Claude Code has quickly gained popularity among software engineers, leading to a substantial increase in the company's revenue, with this five-year-old startup's valuation recently surpassing OpenAI for the first time. In response, OpenAI has prioritized its own programming tool Codex for development.

However, corporate clients' enthusiasm for AI spending is cooling. Some companies that had made significant investments in Anthropic's products have begun seeking to control spending. Earlier this year, an Uber executive indicated that the company’s budget for autonomous AI usage for 2026 has already been exhausted; another company leader mentioned last month that it is challenging to directly link improvements in AI programming efficiency to the rollout of new customer features.

This sentiment has sparked widespread discussions in Silicon Valley about "tokenmaxxing"—the practice of using as many Tokens as possible without regard to returns to enhance productivity.

Price War Tests Business Models, IPO Prospects Add Uncertainty

A price war will serve as an early stress test for the business models of both companies, as they are both on the verge of highly anticipated IPOs. OpenAI and Anthropic currently dominate the emerging AI product revenue space, supporting their rapid growth. However, a potential risk identified by investors over time is the high substitutability of the two companies' products, with relatively low switching costs for customers moving from one to the other.

This week, OpenAI has secretly submitted its IPO application, following in the footsteps of Anthropic. Sam Altman indicated in a recent Slack message to employees that the company plans to go public "within the next year."

OpenAI stated in its confidential filing that "some things may be easier to do as a private company," but did not elaborate further. Against this backdrop, whether a price reduction strategy can effectively increase market share without further widening losses will be a key issue for investors to scrutinize when assessing its IPO value.