NVIDIA is reportedly in discussions to join a funding round for OpenAI that could value the AI startup at more than $100 billion, according to the Wall Street Journal. In May 2024, OpenAI was valued at approximately $86 billion.
This news comes on the heels of NVIDIA’s impressive July quarter results, where revenues surpassed $30 billion—a 122% increase from the previous year.
Besides NVIDIA, Apple and Microsoft are also considering participating in the financing. Thrive Capital is reportedly leading the round with a $1 billion investment, while NVIDIA is evaluating a potential contribution of around $100 million, added the report. Notably, Microsoft has invested $13 billion in OpenAI overall.
While OpenAI is dependent on the NVIDIA GPUs to train its upcoming frontier model, Apple recently partnered with the company, integrating ChatGPT into Siri.
On Thursday, it was reported that ChatGPT has surpassed 200 million weekly active users, doubling its count from the previous year.
Surprisingly, this year, OpenAI has released only GPT-40 and GPT-40 Mini. However, the company has announced several other products, including Sora, SearchGPT, Voice Engine, GPT-40 Voice, and most recently, Strawberry and Orion. It seems that the announcements were likely intended to generate hype and raise funds.
NVIDIA is Investing in Itself, Not OpenAI
Following the investment, the money will ultimately flow back to NVIDIA as OpenAI purchases more compute resources to train its next frontier model.
NVIDIA is keen to secure its ecosystem for the year ahead and is now concentrating on its Blackwell GPUs. This lineup includes models B100 and B200, built for data centres and AI applications.
NVIDIA chief Jensen Huang said that the Blackwell is expected to come out by the fourth quarter this year. “We’re sampling functional samples of Blackwell, Grace Blackwell, and a variety of system configurations as we speak. There are something like 100 different types of Blackwell-based systems that were shown at Computex, and we’re enabling our ecosystem to start sampling those,” said Huang.
However, previous reports indicated that these could be delayed by three months or more for Blackwell due to design flaws, a setback that could affect customers such as Meta Platforms, Google, and Microsoft, which have collectively ordered tens of billions of dollars’ worth of these chips.
Huang believes this is just the beginning and that there’s much more to come in generative AI. “Chatbots, coding AIs, and image generators are growing rapidly, but they’re just the tip of the iceberg. Internet services are deploying generative AI for large-scale recommenders, ad targeting, and search systems,” he said.
According to the NVIDIA CFO Colette Kress the next-generation models will require 10 to 20 times more compute to train with significantly more data.
Earlier this year, Huang personally hand-delivered the first NVIDIA DGX H200 to OpenAI.
OpenAI’s GPT-40 voice features, demonstrated during the Spring Update event, were made possible with the help of NVIDIA H200. “I just want to thank the incredible OpenAI team, and a special thanks to Jensen and the NVIDIA team for bringing us the advanced GPU that made this demo possible today,” said OpenAI CTO Mira Murati, during the OpenAI’s Spring Update.
Apple Wants a Slice of OpenAI
Apple is catching up in the AI race. The company recently released iOS 18.1 beta 3, introducing the AI-powered Clean Up tool under Apple Intelligence, which removes unwanted objects from photos to enhance image quality.
Apple Intelligence’s Clean Up is quite good! So easy to use too. Really really good object segmentation. pic.twitter.com/IjgQ0u4jgs
— Dylan (@DylanMcD8) August 29, 2024
This feature of Apple Intelligence is based on the 3 billion parameter model, which Apple developed recently.
While Apple Intelligence is perfect for day to day tasks, it is not focusing on better reasoning capabilities which will be required in the near future. This is where OpenAI comes into the picture.
“This is a sign that Apple is not seeing a path where it makes sense to build a competitive, full feature LLM,” said Gene Munster, Managing Partner, Deepwater Asset Management on Apple’s investment in OpenAI.
He added that this means Apple will be reliant on OpenAI, Google, and possibly even Meta to deliver about a third of their AI features in the long term.
OpenAI chief Altman is a huge fan of Apple, and his startup eventually ended up partnering with the company. He recently lauded the Cupertino-based tech giant for its technology prowess, saying, “iPhone is the greatest piece of technology humanity has ever made”, and it’s tough to get beyond it as “the bar is quite high.”
As a part of Apple-OpenAI partnership, iOS, iPadOS, and macOS users would get access to ChatGPT powered by GPT-4o later this year, where users can access it for free without creating an account, and ChatGPT subscribers can connect their accounts and access paid features right from these experiences.
Interestingly, when OpenAI announced the ChatGPT desktop app, it was first released for Mac users rather than for Microsoft.
Moreover, it was said that the company wasn’t paying OpenAI anything, as it was doing the startup a favour by making ChatGPT available to billions of customers.
However, investing in OpenAI today would be a smart move for Apple, as it would provide access to the latest OpenAI models, similar to how Microsoft’s AI services primarily rely on OpenAI.
Meanwhile, OpenAI definitely has a soft corner for Apple. This affinity was clearly displayed at the OpenAI Spring Update, where MacBooks and iPhones were prominently used, while Microsoft Windows products were notably absent.
The post NVIDIA, Apple Blow OpenAI’s Bubble appeared first on AIM.
