Chip startup Delos Data, founded by former Intel engineers, announced a $100 million financing round aimed at solving the data‑center bottleneck that current GPUs struggle to address. The capital will be used to develop specialized processors optimized for high‑throughput AI workloads, reducing latency and energy consumption in large‑scale model training.
Industry analysts see the move as a response to growing demand for more efficient AI infrastructure as generative models expand in size. If successful, Delos’s technology could lower operating costs for cloud providers and accelerate the rollout of next‑generation AI services, reshaping the competitive landscape between traditional GPU vendors and emerging ASIC players.
This story was written by a language model from reporting by TechStartups.com. Check the original before quoting it.