Nvidia set to launch new chip that could reset the AI race, says report — Key things to know
In a major change that is likely to reset the AI race, Nvidia plans to launch a new processor as early as next month that will drastically overhaul the way its customers use the technology, the Wall Street Journal reported, citing sources with knowledge of the matter.
The new processor is specifically designed to help OpenAI and other customers build faster, more efficient tools, the report said.
Here’s everything you need to know about Nvidia’s new processor.
Things to know about Nvidia’s new processor
Nvidia is in the process of designing a new system for “inference” computing. This will allow the company’s customers and AI models to respond to queries, according to the WSJ.
According to the report based on sources, the new platform will be introduced at Nvidia’s GTC developer conference to be held in San Jose in March. It will also include a chip designed by startup Groq.
Inferential computing has emerged as a key component of the intense AI race the world is currently seeing. Nvidia rivals Google and Amazon have designed chips that can compete with the company’s current flagship systems. But now, the explosion of autonomous coding in the tech workforce has triggered the need for new chips that can efficiently handle complex AI-related tasks.
The report also stated that ChatGPT maker OpenAI has agreed to become one of the largest customers of Nvidia’s new system. Earlier this month, Reuters separately reported that OpenAI was unhappy with the speed at which Nvidia’s hardware responded to ChatGPT users for certain types of issues, such as software development and the AI communicating with other software.
In the future, OpenAI needs new hardware that will provide approximately 10% of its inference computing needs. Reuters He stated that he was quoting someone who knew.
This has put Nvidia under pressure as its customers push for inference generation as AI continues to unlock new capabilities every day. Until now, Nvidia has dominated the market with its high-end GPUs, which are enough for its customers to run AI operations. Most analysts estimate that Nvidia controls 90% or more of the GPU market.
But earlier Friday, OpenAI announced it would look to buy “dedicated inference capacity” from Nvidia, possibly pointing to the new processor. Nvidia will also invest $30 billion in ChatGPT maker. OpenAI also signed a major new deal in which it will use Amazon’s Trainium chips.
The ChatGPT maker has discussed working with startups like Cerebras and Groq to provide chips for faster inference, Reuters reported, citing two people with knowledge. However, according to the report, Nvidia made a $20 billion licensing agreement with artificial intelligence startup Groq, which stopped OpenAI’s negotiations.
In September, Nvidia said it planned to pour as much as $100 billion into OpenAI as part of a deal that gives the chipmaker a stake in the startup and gives OpenAI the money it needs to buy advanced chips.
