Jensen Huang: The $30 billion investment in OpenAI "may be the last time" OpenAI is expected to launch an IPO by the end of this year

NVIDIA CEO Jensen Huang stated that the company’s recent $30 billion investment in OpenAI may be its last before the company goes public. He expects OpenAI to launch an IPO by the end of this year.

At the Morgan Stanley Technology, Media, and Telecom Conference on Wednesday, Huang claimed that the infrastructure deal involving $100 billion, disclosed by both companies last September, may “not materialize.” “The reason is that they are about to go public.”

He also mentioned that NVIDIA previously announced a $10 billion investment in OpenAI’s competitor, Anthropic, which “likely will also be the last.” NVIDIA disclosed this investment plan when it jointly announced with Microsoft in November last year.

Huang’s comments come after months of speculation about the scale of NVIDIA’s collaboration with OpenAI. The chipmaker disclosed in a quarterly filing last November that the previously announced $100 billion deal might not be completed, and media reports in January indicated the agreement had been “shelved.”

In NVIDIA’s February quarterly earnings report, similar language was used, stating that the company “cannot guarantee” reaching an “investment and partnership agreement” with OpenAI or that the deal would be finalized.

NVIDIA’s $30 billion investment in OpenAI is part of the $110 billion funding round announced last Friday. The round also includes $50 billion from Amazon and $30 billion from SoftBank, valuing OpenAI at $730 billion pre-money.

According to the agreement, OpenAI received 3 gigawatts of dedicated inference compute power and 2 gigawatts of training compute power from NVIDIA’s Vera Rubin AI data center system.

Notably, the agreement reached between the two companies in September last year shocked the tech industry and triggered a series of subsequent infrastructure cooperation agreements. The deal outlined NVIDIA’s gradual investment in OpenAI over the coming years as new supercomputing facilities come online. However, this $30 billion investment is not tied to specific deployment milestones.

NVIDIA has been one of the biggest beneficiaries of the AI boom, thanks to its GPUs used for training AI models and running large-scale workloads.

Nevertheless, as AI demand shifts from training to inference, NVIDIA faces some pressure. Reports indicate that NVIDIA is developing a new chip specifically for inference, and OpenAI is expected to be one of the largest customers for this new processor.

In February, OpenAI announced it would purchase substantial “dedicated inference compute” from NVIDIA. OpenAI is also heavily investing in Amazon’s inference-optimized chips and using Google’s TPU.