Safe Superintelligence Secured $8 Billion Funding

The AI startup reached a $32 billion valuation after Nvidia committed $5 billion in equity to the venture.

Updated on Sept. 18, 2026 in Artificial Intelligence

Isometric editorial illustration of a clean, geometric server facility stack, representing large-scale corporate investment in AI infrastructure. — Safe Superintelligence secured $8 billion in total funding on July 27, 2026, reaching a $32 billion valuation led by a $5 billion equity stake from Nvidia. AI Illustration. Upload story photo >

Live Poll

Is now a good time to invest in companies that have not yet launched a product?

On July 27, 2026, Safe Superintelligence successfully raised $8 billion in total funding, bolstered by a $5 billion equity investment from Nvidia. The company, which has not yet released any products, achieved a $32 billion valuation.

Why it matters

Ilya Sutskever, who left OpenAI in 2024, founded the firm on the premise that the industry scaling era is over. The startup focuses exclusively on developing safe superintelligence as its sole product destination.

Nvidia’s $5 billion investment grants Safe Superintelligence priority access to Vera Rubin compute platforms. This partnership is projected to increase the startup's available compute capacity tenfold over the next year.

The players

Safe Superintelligence

This startup is led by former OpenAI executive Ilya Sutskever and focuses on developing safe superintelligence.

Nvidia

The technology corporation provides hardware and compute platforms for advanced artificial intelligence research.

Ilya Sutskever

He is the founder of Safe Superintelligence and previously served as a high-ranking official at OpenAI until 2024.

Gavin Baker

He is an investor who publicly commented on the company's development roadmap in August 2026.

The details

Safe Superintelligence uses its capital as fuel to reach its singular unreleased product goal. While investor Gavin Baker claimed in August 2026 that the firm planned a model release, the company continues to operate without any publicly available software.

Ilya Sutskever departed from OpenAI in 2024.
The firm’s founder discussed AI eras on a podcast in November 2025.
Nvidia announced its $5 billion investment on July 27, 2026.
Investor Gavin Baker speculated on a potential model release in August 2026.

The Tech Race

This deal underscores the deepening reliance on Nvidia Vera Rubin compute platforms as the critical bottleneck for AI development. It signals a move away from standard large-model scaling toward specialized, resource-heavy architectures in the race for superintelligence.

While the company has not yet released consumer-facing products, the massive influx of compute resources suggests accelerated development of advanced AI models. Users should monitor this space to see if the firm's focus on safety shifts how future AI systems are deployed in real-world software.

The takeaway

The massive investment suggests that industry leaders believe the future of AI lies in proprietary, compute-intensive innovation rather than incremental model scaling. This capital-heavy approach highlights how expensive and hardware-dependent the pursuit of superintelligence has become.