Tech Giants Invest in AI Startup Safe Superintelligence, Signaling New Era of Collaboration

SAN FRANCISCO: In a significant development within the artificial intelligence landscape, Alphabet Inc. and Nvidia have teamed up with several notable venture capital investors to support an ambitious startup known as Safe Superintelligence (SSI). This innovative company was co-founded by Ilya Sutskever, the former chief scientist at OpenAI, and has quickly emerged as one of the most valuable AI startups since its inception just months ago, according to sources acquainted with the deal.
The funding round highlights a renewed interest from major technology companies and infrastructure providers in making strategic investments in startups that are pushing the boundaries of advanced AI technologies, particularly those demanding substantial computing resources. Earlier this week, Alphabet's cloud computing division made headlines by announcing an agreement to provide SSI with access to its proprietary tensor processing units (TPUs)the company's specialized AI chips known for their efficiency in handling complex computations.
SSI has garnered significant attention and financial backing, with reports indicating that the startup was recently valued at an impressive $32 billion in a funding round led by Greenoaks Capital. This valuation underscores SSI's position as one of the leading players in the AI model research sector, bolstered by Sutskever's exceptional track record in identifying and developing the next breakthroughs in artificial intelligence.
As demand for AI chips continues to escalate, both Alphabet and Nvidia find themselves in a race to capture more market share. However, the precise terms of the investments made by Alphabet and Nvidia in SSI remain undisclosed, as representatives from all three companies have opted not to comment on the financial specifics of the deal.
These strategic moves by Alphabet's corporate and cloud divisions, alongside high-profile AI labs such as SSI and Anthropic, reveal the tech giant's evolving strategy regarding AI hardware. Initially, Google reserved its TPUs solely for internal projects; however, the recent agreement to offer these chips to SSI in significant quantities demonstrates a broader shift towards marketing its hardware to external customers. Darren Mowry, a managing director overseeing Google's partnerships with startups, emphasized this shift in an interview with Reuters, stating, With these foundational model builders, the gravity is increasing dramatically over to us.
Historically, AI developers have favored Nvidia's graphics processing units (GPUs), which command over 80 percent of the AI chips market. However, it appears that SSI is primarily utilizing TPUs, rather than GPUs, in its AI research and development processes, as reported by two informed sources.
Google's cloud service offers both Nvidia GPUs and its own TPUs, with the latter designed to excel in specific AI tasks and outperform general-purpose GPUs in terms of efficiency. These TPUs have been instrumental in constructing large-scale AI models, similar to those developed by competitors like Apple and Anthropic, the latter of which has attracted billions of dollars in funding from both Google and Amazon.
Adding another layer of competition to the landscape, Amazon has been developing its own AI processors known as Trainium and Inferentia. Back in 2023, Amazon announced that Anthropic would leverage these chips for their technological advancements. In December, Amazon also revealed plans for Anthropic to become the first client of its colossal supercomputer, powered by hundreds of thousands of its proprietary chips.
Despite this competitive environment, Anthropic has continued to rely on TPUs for its AI development without reducing its expenditures on Google's chips, as indicated by two sources familiar with the matter.
As major cloud providers increasingly invest in AI startups that not only build foundational models but also become significant customers for their infrastructure, the dynamics of the tech industry are shifting. Companies like Amazon and Google have both made substantial investments in Anthropic, while Microsoft has placed significant bets on OpenAI. Recognizing the potential of these startups, Nvidia has similarly backed OpenAI as well as Elon Musk's xAI, showcasing the collaborative spirit within this rapidly evolving field.