OpenAI Explains Why It's Slashing AI Prices — Michael Burry Says The 'Real News' Is Preparation For DeepSeek's New V4 Models

• OpenAI said its goal is not simply to build larger models or more data centers, but to deliver “more useful intelligence within reach.”
• According to a Nikkei Asia report, DeepSeek on Friday made the official beta version of its V4 Flash model publicly available.
• Burry said OpenAI's recent AI price cuts were aimed at preparing for DeepSeek's V4 model launch.

OpenAI on Friday explained the rationale behind the latest price cuts for its AI models, saying lower costs are central to its strategy of making artificial intelligence more widely accessible.

The Microsoft Corp. (MSFT) backed company argued that AI infrastructure is valuable not because of its scale, but because it enables more capable intelligence at a lower cost.

OpenAI said falling prices expand the range of tasks customers can economically automate, generating more demand and helping fund the next generation of models.

The company also announced that its models are now used by over a billion active users and more than two million businesses.

OpenAI Says Lower AI Prices Expand Adoption

OpenAI said its goal is not simply to build larger models or more data centers, but to deliver “more useful intelligence within reach.”

According to the company, lower prices make more AI workloads economically viable, while improvements in model capability and efficiency create a feedback loop that drives broader adoption and supports further investment.

OpenAI also highlighted recent efficiency gains, saying GPT-5.6 Sol helped reduce end-to-end serving costs by 20% by optimizing production software and improving speculative decoding to increase token-generation efficiency by more than 15%.

It added that advances in routing, context management, and system design can significantly improve performance without changing the underlying model.

DeepSeek's V4 Rollout Draws Michael Burry's Attention

According to a Nikkei Asia report, DeepSeek on Friday made the official beta version of its V4 Flash model publicly available. The company said that the updated model delivers significantly stronger agent capabilities than its preview release. The company also said the official version of V4 Pro will be released soon.

The launch comes as competition in AI model pricing continues to intensify. DeepSeek announced a permanent 75% price cut in May 2026 for its V4 model services.

“The Big Short” investor Michael Burry reacted to DeepSeek's beta release of its V4 models, stating in a post on X that “the real news is OpenAI slashing and burning its prices to prepare for this,” suggesting the company's recent pricing moves reflect intensifying competition in the AI model market.

On Thursday, OpenAI cut the price of the GPT-5.6 Luna model by 80% and GPT-5.6 Terra by 20%, while introducing a faster processing option for GPT-5.6 Sol.

OpenAI said that the pricing changes were designed to expand the range of work customers can perform by giving them greater flexibility to balance intelligence, speed, reliability and cost.

Earlier this month, OpenAI Chairman Bret Taylor downplayed the cost advantage that open-weight models offer users, saying that token efficiency matters more.

Taylor argued that the cost of training open-weight models is less important than how efficiently they perform for end users.

According to data from Artificial Analysis, DeepSeek V4 Flash 0731 (max) costs $0.03 per Intelligence Index Task, while OpenAI’s GPT 5.6 Sol (max) costs $1.86 and Anthropic’s Claude Fable 5 (with fallback) costs $3.15.

The iShares U.S. Technology ETF (IYW) is up 30% year-to-date, while the Global X Artificial Intelligence & Technology ETF (AIQ) is up 31%.

Also See: AXTI Stock Skyrockets As Investors Cheer Blowout Guidance, Record Quarterly Revenue — Needham Spots A 'Real' AI Networking Opportunity

For updates and corrections, email newsroom[at]stocktwits[dot]com.