AMD announced its next-generation AI chips, the Instinct MI400 series, set to launch next year, at an event in San Jose. CEO Lisa Su highlighted the new “rack-scale” Helios system, designed to integrate thousands of MI400 chips into unified compute clusters for large data centers. OpenAI CEO Sam Altman appeared alongside Su, confirming OpenAI’s plans to use the new AMD chips. The MI400 chips are positioned as a direct rival to Nvidia’s leading GPU offerings, with AMD emphasizing lower operating costs, aggressive pricing, and superior inference capabilities. Major technology companies, including Microsoft, Oracle, Meta, and Tesla, are adopting or planning large-scale deployments of AMD’s AI chips, with AMD striving to capture a greater share of the rapidly expanding AI hardware market, currently dominated by Nvidia.
Related articles:
Tesla expands AI infrastructure for autonomous vehicles
AI startups turn to AMD for GPU power amid chip shortage
