Nvidia AI chip rivals secure record funding amid rising competition
Nvidia’s graphics processing units have become essential for training AI models
Nvidia remains at the center of the global artificial intelligence (AI) boom, but a growing wave of startups is drawing billions in funding as investors bet on alternatives to the chip giant’s dominance.
According to data from Dealroom, AI chip startups raised $8.3 billion globally in 2026, with funding expected to climb further barring a market downturn.
The surge reflects increasing interest in new chip architectures designed specifically for AI workloads.
Nvidia’s graphics processing units (GPUs), originally built for gaming, have become essential for training AI models.
However, the industry is shifting focus toward “inference” — the process of deploying AI in real-world applications — where startups argue current GPU designs are less efficient.
“Inference has taken the lead now, and the current GPU design isn’t optimized for large-scale applications,” said Patrick Schneider-Sikorsky of the NATO Innovation Fund, an investor in UK-based startup Fractile.
Despite rising competition, Nvidia continues to invest heavily. The company spent over $18 billion on research and development in its latest fiscal year and has expanded through acquisitions and investments in emerging technologies.
Meanwhile, startups are securing major backing. In the US, firms like Cerebras Systems raised $1 billion, while companies including MatX, Ayar Labs, and Etched attracted $500 million rounds. In Europe, Axelera and others have also raised significant capital.
Industry leaders say AI chips are no longer a niche investment but a core component of future infrastructure. “This is evolving into a fundamental part of AI planning,” said Carlos Espinal of Seedcamp.
The broader AI ecosystem is also expanding. OpenAI and Anthropic are growing their presence in the UK, while TSMC reported a 58% jump in quarterly profit driven by strong AI chip demand.
Even as competition intensifies, Nvidia’s position remains strong — but the race to redefine AI hardware is clearly accelerating.