Following Google Gemini Large Model Price Cuts, OpenAI Also Set for Price Reduction Wave; Will It Evolve Into ‘New Energy Vehicle-Style’ Competition?

TradingKey - On June 10, ET, according to the Wall Street Journal, OpenAI is considering significantly reducing the fees charged to users (i.e., token pricing) in an effort to win customers from its main competitor, Anthropic.

Sources familiar with the matter revealed that the move is in response to similar price cuts expected from Anthropic. OpenAI anticipates that competition in the AI services market will further escalate and potentially evolve into a cost war.

Previously, OpenAI launched the Flex Processing API option, offering a 50% discount on standard prices in exchange for compromises on response speed and stability: the per-million input tokens for the o3 model in Flex mode dropped from $10 to $5, and the per-million output tokens dropped from $40 to $20.

However, the latest WSJ report suggests that the new round of price cuts being considered could be more aggressive and may have an expanded scope.

This news has pushed the price war of large AI models to the forefront. This is highly similar to the industry-wide "involution" triggered by price cuts in the new energy vehicle (NEV) industry in recent years. As technological gaps narrow and moats shallow, price adjustments become the final move to reshape the industry landscape.

The core reason why NEVs engage in mutually destructive competition is severe product homogenization and low customer loyalty. Compared to the NEV battlefield, the greatest danger of the AI price war lies in its prisoner's dilemma structure.

Due to the high interchangeability of AI products, enterprise customers can easily switch API interfaces between various large models based on price.

The WSJ report noted that due to the high interchangeability of AI products and the ease of switching for enterprise customers, any first-mover price cut could accelerate a broad repricing of AI services, forcing competitors to follow suit quickly. The result would not be a winner-take-all scenario, but a general compression of profit margins across the industry, posing a fundamental challenge to the commercial sustainability of large-model companies that trade compute for revenue.

From a global enterprise perspective, Google has previously lowered Gemini model API prices several times and launched lightweight models, while Chinese vendors such as DeepSeek have slashed API prices to "nominal" levels.

If OpenAI and Anthropic both implement significant price cuts, it is expected to fully ignite a global price war for large AI models.

The market is concerned that the wave of price cuts for large AI models may evolve from a technical race into a vicious competition of burning cash to survive. While price cuts can seize developer share in the short term, the path to commercialization will be continuously squeezed, and the profit logic of the entire industry may face reshaping.

Wall Street Insights There is also a view that the possibility of truly significant price cuts by Anthropic and OpenAI is not high, given the tangible costs of computing power, storage, networking, and electricity behind AI services. More importantly, B2B subscriptions have not seen substantial problems so far. The so-called significant price cuts are more likely to occur in the B2C segment, aimed at competing for a larger user base, higher usage frequency, and a stronger ecosystem entry point.