OpenAI Will Use Rival's Chips for Artificial Intelligence Infrastructure

OpenAI plans to purchase rival Google's Tensor Processing Unit (TPU) units, which are artificial intelligence chips, to train and run artificial intelligence models. Details about this possible agreement, such as the financial size, delivery schedule or the amount of chips to be purchased, have not yet been shared.

Commercial cooperation between the two companies has been increasing recently despite the competition in the fields of artificial intelligence products and search engines. OpenAI announced last week that it would offer the most advanced artificial intelligence models via Google Cloud.

If the new chip agreement is realized, this cooperation between the companies will also be transferred to the hardware side. OpenAI is among the largest customers currently using Nvidia's graphics processing units (GPU) to train and run AI models.

The company aims to diversify by turning to alternative solutions to Nvidia to meet its hardware needs. Google's custom processors, known as TPUs, are gaining increasing attention in the industry as an alternative to Nvidia's GPUs.

Companies are evaluating different hardware options to reduce the cost of artificial intelligence infrastructure. Google also started selling TPU chips directly to some customers this year.

Demand for Google's chips is soaring thanks to deals with rival AI companies like Anthropic. Income in this area is expected to reach a significant level, especially in 2027.

OpenAI is also working on its own hardware solutions in order to reduce its dependence on external suppliers. The company introduced its first special artificial intelligence chip, Jalapeno, which it designed together with Broadcom, in June.

OpenAI aims to put this processor, which is part of a multi-generation chip program, into use by the end of 2026. The company's development of its own chip is considered as part of its long-term hardware independence strategy.

OpenAI's move towards Google chips coincides with a period when major investments in artificial intelligence infrastructure in the technology sector are accelerating. Alphabet, the parent company of Amazon, Microsoft and Google, is expected to spend up to 600 billion dollars in total on artificial intelligence infrastructure in 2026.

Investors continue to question when these massive escalating expenses will generate meaningful financial returns for companies. The quarterly financial results to be announced by Alphabet are closely followed by investors who want to see the contribution of artificial intelligence investments to the cloud computing business line.

This competition in the cloud computing market directly affects the infrastructure investments required for artificial intelligence models to work more efficiently. These moves of companies regarding hardware supply are among the fundamental factors that will shape the future capacity of the sector.

These huge expenditures for artificial intelligence infrastructure also intensify the competition between hardware suppliers. In the technology world, hardware diversity continues to play a critical role in the development processes of artificial intelligence models.