Nvidia has invested in Ilya Sutskever's AI startup. Why is it doing this?

Nvidia has invested a “significant” amount in Safe Superintelligence (SSI), a startup founded by Ilya Sutskever, the former chief scientist at OpenAI. In addition to funding, the startup will gain access to Nvidia’s new Vera Rubin computing platform, according to a joint statement released by the companies on July 27.

The companies did not disclose the financial terms of the deal. According to sources cited by the Financial Times, the chipmaker agreed to invest approximately $5 billion in Safe Superintelligence. Such large investments typically depend on the startup meeting certain milestones, the publication notes.

Ilya Sutskever was born in the former Soviet Union and grew up in Israel. The Wall Street Journal calls him a mathematical prodigy. Sutskever rose to prominence while pursuing his graduate studies in Canada under the guidance of Turing Award laureate and leading artificial intelligence researcher Geoffrey Hinton. In 2012, Sutskever and Hinton co-authored a paper on computer vision that sparked the modern race in the field of AI. The paper demonstrated that scaling deep neural networks using vast amounts of data and computational power leads to a significant increase in their intellectual capabilities.

Sutskever then joined Google, but left in 2015 to become one of the co-founders of OpenAI. It was there that his ideas played a key role in the creation of ChatGPT, according to the WSJ. At OpenAI, Sutskever led the Superalignment team, which worked to ensure that the AI created by the company would not harm humanity. According to The Wall Street Journal, in November 2023, it was Sutskever who initiated Sam Altman’s brief removal from the position of CEO of OpenAI. The reason for their conflict, according to WSJ sources, may have been disagreements over AI safety: Sutskever was reportedly concerned that AI would eventually treat humans the same way humans currently treat animals. Before Altman was reinstated, Sutskever issued a public apology, stating that he regretted his involvement in the campaign.

Shortly after leaving OpenAI in 2024, Sutskever co-founded Safe Superintelligence. The startup’s goal is to create safe artificial intelligence systems. Safe Superintelligence says it has no plans to sell AI products in the near future, according to Bloomberg. In July, Sutskever announced that he would lead the company as CEO—after the previous CEO and fellow co-founder Daniel Gross left to join Meta.

Safe Superintelligence has raised approximately $3 billion from various investors, including venture capital firms Andreessen Horowitz, Sequoia Capital, and DST Global. Last year, the startup was valued at $32 billion.

The deal with SSI reflects Nvidia’s overall strategy: the company actively supports cutting-edge AI startups that also purchase its chips, according to Bloomberg. Such deals are often criticized for being circular, as they create a mutual dependency between the hardware supplier and AI developers, the agency notes. At the same time, according to the WSJ, Nvidia is discussing providing a guarantee of up to $250 billion. This funding is intended to help OpenAI lease computing power as part of a U.S. project to build data centers.

This article was AI-translated and verified by a human editor