Nvidia's grip on AI chips weakens as Microsoft turns to AMD and Anthropic may follow
Quick Answer
Microsoft is shifting to AMD's Helios AI chips for Azure, aiming to diversify from Nvidia's GPUs.
Quick Take
AMD's partnership with Anthropic is also in the works, enhancing its competitive stance in AI hardware.
Key Points
- Microsoft expands partnership with AMD for Helios AI chips in Azure cloud.
- Helios platform designed for large AI models, shipping in late 2026.
- AMD prioritizes Anthropic as a potential major customer for AI hardware.
- Nvidia faces increased competition from AMD and custom TPUs.
- OpenAI also collaborates with AMD for AI infrastructure.
📖 Reader Mode
~1 min read
Jul 20, 2026
Microsoft plans to use more AMD AI chips in its Azure cloud. The expanded partnership centers on AMD's new Helios platform, which is scheduled to ship in the second half of 2026. Helios is designed to run large AI models and give Azure customers an alternative to Nvidia's dominant GPU systems. AMD CEO Lisa Su called the partnership a major milestone. Microsoft CEO Satya Nadella said customers need more choices for AI infrastructure. Nvidia still has considerable pricing power, but it faces growing competition from AMD and custom TPUs made by other companies.
AMD could also land another major customer. A public GitHub profile belonging to an AMD director suggests that Anthropic, the company behind Claude, is testing AMD hardware. According to SemiAnalysis, AMD gave Anthropic its highest priority rating, putting it on par with major customers such as Meta. Anthropic is still testing the hardware. AMD may confirm the partnership at its upcoming "Advancing AI" conference if its engineering team resolves the remaining concerns about software quality. OpenAI also has a partnership with AMD.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
