Enterprises put non-Nvidia chips 14 points ahead of Nvidia's next-gen GPUs on their evaluation lists
According to a July survey by VentureBeat of 170 AI infrastructure respondents, 39.4% of enterprise buyers are likely to include non-NVIDIA chips (such as AWS Trainium, Google TPU, AMD Instinct, Intel Gaudi, or self-developed ASICs) in their evaluation lists for building the next-generation AI accelerators, which is 14 percentage points higher than the 25.3% for NVIDIA Blackwell (GB300) or other new-generation NVIDIA GPUs. Although NVIDIA remains the default choice for most production environments, organizations are optimizing their existing AI infrastructure by introducing real options rather than considering NVIDIA as the only evaluation criterion. Microsoft Azure saw the most significant growth in production deployment, with the proportion of respondents rising from 29% in June to 47.1% in July; Google Gemini and OpenAI also experienced increases in production adoption rates of 6.5 and …