Earlier this month, it just spent a whopping $20 billion on AI chips . Shortly after startup Groq reached a "non-exclusive licensing agreement," Nvidia... It seems Nvidia has no intention of stopping its acquisition spree. According to local Israeli sources, Nvidia is reportedly considering acquiring an Israeli artificial intelligence company for up to $3 billion. The startup AI21 Labs is in advanced negotiations.
AI21 was valued at $1.4 billion in its most recent funding round in 2023. Nvidia and Google's parent company, Alphabet, both participated in that round.
Founded in 2017 by Amnon Shashua and two others, AI21 is one of the startups that benefited from the artificial intelligence boom, attracting significant interest from venture capital firms and other investors. Notably, Shashua is also the founder and CEO of Mobileye, a developer of autonomous driving technology.
According to local media reports, AI21 has long been "for sale," and recent negotiations with Nvidia have made significant progress. Nvidia's main interest in AI21 Labs appears to lie in its workforce of approximately 200 employees—most of whom hold advanced degrees and possess "rare expertise in the field of artificial intelligence development."
It is understood that the transaction amount for Nvidia's acquisition of AI21 is expected to be between $2 billion and $3 billion.
As the world's most valuable company, Nvidia is currently planning a major expansion in Israel – it will build a new R&D campus in Kiryat Tivon, south of the port city of Haifa (Israel's third largest city), which can accommodate up to 10,000 employees.
Nvidia CEO Jensen Huang has described Israel as the company's "second home." Nvidia stated that the campus, once completed, will cover 90 Dunhams (22 acres) and include up to 160,000 square meters (1.7 million square feet) of office space, parks, and public areas. Its design is inspired by Nvidia's headquarters in Santa Clara, California. Nvidia expects construction to begin in 2027, with the first office space planned for operation in 2031.
For Nvidia, if this acquisition is successful, it will be its fourth major acquisition in Israel, and its second largest acquisition after the $7 billion purchase of Mellanox. In 2023, Nvidia also acquired Deci and Run:ai for a total of $1 billion.
In recent years, Nvidia, with its deep pockets, has been continuously leveraging its financial power for expansion. Just last week, Groq, once considered one of Nvidia's challengers, announced on its official website that it had reached a non-exclusive licensing agreement with Nvidia. Groq founder and CEO Jonathan Ross, President Sunny Madra, and other core executives and teams will join Nvidia.
Given the non-exclusive licensing agreement between NVIDIA and Groq, NVIDIA will acquire Groq's core AI inference technology intellectual property rights. The rumored acquisition of AI21 Labs seems to primarily target its workforce. It's clear that Nvidia is currently striving to maintain its absolute dominance in the AI chip field through a combination of strategies: consolidating the CUDA ecosystem, leveraging multi-architecture AI computing power, and attracting more AI chip talent.
For the past two years, AI21 has struggled to keep pace with the rapid development of industry leaders. In April of this year, the company discontinued development of Wordtune, its consumer-facing AI-assisted writing and reading product. Now, AI21 primarily focuses on professional language models for enterprise clients, which have extremely high requirements for accuracy and reliability and a much lower tolerance for errors compared to consumer applications.
The company's flagship enterprise product, Maestro, aims to improve the accuracy of large language models by up to 50%. In recent months, AI21 has also launched a new inference model that is claimed to be faster, more efficient, and less memory-intensive than competing systems. Nevertheless, the company's annual revenue is estimated at around $50 million, far less than the billions of dollars in revenue earned by its leading AI competitors.
(Article source: CLS)
