The partnership was the latest in a series of recent announcements by OpenAI as it seeks to strengthen its position as the preeminent company of the generative AI revolution that began with the release of ChatGPT in November 2022.

In the past few weeks, under the leadership of CEO Sam Altman, OpenAI has signed huge investments in data centers and AI chips with US companies Nvidia and AMD, as well as with South Korea's Samsung and SK Hynix.

The partnerships come even as there are no real signs that the AI business is close to breaking even, despite its strong growth.

The deals to build out AI's immense infrastructure needs also threaten further stress on electricity providers, with AI data centers demanding significant power and resources to deliver the computing necessary for the expected usage of the technology.

The financial terms of the deal were not part of the announcements and remain unclear for many of the recent deals, drawing some skepticism from observers who worry the AI frenzy may have created a financial bubble, posing a risk to investors.

By designing its own chips rather than relying solely on off-the-shelf processors, OpenAI says it can build hardware specifically tailored to how its AI models work, potentially making them faster and more powerful.

"Partnering with Broadcom is a critical step in building the infrastructure needed to unlock AI's potential and deliver real benefits for people and businesses," said Altman, OpenAI's co-founder and chief executive.

The custom processors will be installed in data centers operated by OpenAI and its partners to keep up with soaring worldwide demand for AI services.

OpenAI has grown rapidly to more than 800 million people using its services weekly since its 2022 launch.

Broadcom, which will manufacture and help develop the chips, called the collaboration "a pivotal moment" in AI development.

arp/aha

BROADCOM

NVIDIA

AMD - ADVANCED MICRO DEVICES

Samsung Electronics