Experts warn acquisition could give chip giant deeper influence over how global AI computing systems operate.
Concerns are mounting among artificial intelligence researchers and supercomputing specialists after Nvidia acquired software firm SchedMD, a move that could reshape control over critical AI infrastructure.
At the center of the debate is Slurm, an open-source workload management software developed by SchedMD. The platform schedules and distributes computing tasks across large AI clusters and powers a majority of the world’s supercomputers, making it a foundational tool for AI training and scientific research.
Experts fear that Nvidia’s ownership of such a widely used software layer may create subtle competitive advantages. Since Slurm decides how computing resources are allocated, any optimisation favouring Nvidia’s GPUs could influence performance outcomes and eventually purchasing decisions by cloud providers, research institutions, and enterprises.
The concern is not about immediate changes but long-term influence. Researchers argue that control over both AI hardware and the software managing workloads could strengthen Nvidia’s already dominant position in AI computing, potentially disadvantaging rival chipmakers.
The worries also stem from earlier industry experiences where tighter integration between Nvidia software tools and its hardware improved performance within its ecosystem, raising questions about vendor neutrality.
Nvidia has pushed back against these fears, saying Slurm will remain open-source and hardware-agnostic. The company maintains that the acquisition will accelerate development and help scale AI infrastructure as global demand for computing power surges.
Still, the deal underscores a broader industry shift: the battle for AI leadership is increasingly moving beyond chips to the software platforms that control how computing power is used.
