Exclusive: Startup aiming to break Nvidia’s strangehold on AI data center workloads raises $10.25 million

Fortune

by Jeremy Kahn

February 26, 2026

AI-Generated Deep Dive Summary

A London-based startup named Callosum, founded by two Cambridge-trained neuroscientists, is challenging Nvidia's dominance in AI data center workloads with its innovative software that optimizes AI tasks across diverse chip types. The company has raised $10.25 million and is receiving research funding from the U.K. government to develop a "sovereign cloud" infrastructure for AI, reducing dependence on U.S. technology providers. Inspired by neuroscience, Callosum's platform distributes AI workloads across chips like Nvidia GPUs, AMD processors, AWS Trainium, and Cerebras' silicon, aiming to enhance performance and efficiency. The startup's approach is rooted in the idea that intelligence arises from diverse systems working together, unlike the current trend of relying on identical hardware. Co-founders Danyal Akarca and Jascha Achterberg believe their software can deliver twice the accuracy, seven times faster performance, and lower costs compared to traditional methods. This shift aligns with the industry's growing focus on inference workloads, which are expected to account for two-thirds of AI compute by 2026. Callosum's platform is designed to integrate seamlessly across major cloud providers like AWS, Google Cloud, and Microsoft Azure, allowing businesses to optimize their existing setups without significant restructuring. The company targets complex enterprise tasks, where its system can achieve superior results compared to single-model approaches. With a growing list of investors and government support, Callosum aims to become the key software layer in an increasingly fragmented hardware landscape. For businesses, this means potential cost savings, improved performance, and reduced reliance on dominant players like Nvidia. As AI workloads evolve, Callosum's ability to leverage diverse chip technologies positions it as a critical player in shaping the future of AI computing.

Verticals

businessfinance

Originally published on Fortune on 2/26/2026