AMD Gains Ground Against Nvidia in AI Chip Competition

What happened

AMD announced several major wins in the AI chip space during the first half of 2026. The company is reportedly finalizing a deal to supply 2 gigawatts of AI computing capacity to Anthropic, potentially including an equity stake worth up to $5 billion. Microsoft's Azure platform committed to using AMD's new Helios rack-scale platform, complementing existing partnerships with OpenAI and Meta Platforms. Additionally, autonomous driving company Turing is migrating 10% of its AI training workloads from Nvidia to AMD. These developments come as AMD's new accelerators are positioned to potentially exceed Nvidia's performance in certain workloads by year-end. Meanwhile, major tech companies are collectively planning to spend over $500 billion on AI infrastructure in 2026, with Alphabet raising its capital spending forecast to $195-205 billion, Amazon targeting $200 billion, Meta planning $125-145 billion, and Microsoft around $190 billion.

Why it matters

For AMD, these partnership wins represent meaningful progress in challenging Nvidia's dominance in the lucrative AI accelerator market. While Nvidia reported $75 billion in data center revenue in a single quarter and maintains over 80% market share, AMD is expanding its footprint in a market estimated to exceed $200 billion in 2026. The massive capital spending commitments from cloud providers and tech giants-totaling over half a trillion dollars-create significant opportunities for AMD to capture more share without necessarily displacing Nvidia. AMD's stock performance reflects this momentum, rising nearly 158% year-to-date as of late July 2026, significantly outpacing Nvidia's 12% gain over the same period. The company's ability to secure commitments from major AI players like Anthropic, Microsoft, and existing partners OpenAI and Meta demonstrates that customers are diversifying their chip suppliers, reducing concentration risk while AMD benefits from growing overall market demand.

Bigger picture

The AI chip race in the first half of 2026 reveals an expanding market rather than a zero-sum competition. While Nvidia continues to collect the majority of AI infrastructure spending with 92% year-over-year growth in its data center segment reaching $75.2 billion quarterly, AMD's gains illustrate the market's enormous scale and growth trajectory. Major tech companies are not simply redistributing existing budgets-they are dramatically expanding total spending, with some raising capital expenditure forecasts mid-year. However, this spending boom also carries risks. Large customers like Meta are developing custom silicon with Broadcom to reduce dependency on external suppliers, while investors increasingly question whether these massive investments will generate adequate returns. Nvidia's stock pullback of roughly 17% from its May peak despite strong fundamentals suggests market concerns about sustainability. Meanwhile, AMD benefits from being both a credible alternative to Nvidia and a well-positioned participant in the broader AI infrastructure buildout, without bearing the same concentration risk.

What to watch

Monitor whether AMD can convert partnership announcements into meaningful revenue growth that narrows the gap with Nvidia's data center business. Watch for updates on the Anthropic deal finalization and the performance benchmarks of AMD's new accelerators compared to Nvidia's Blackwell platform. Track capital spending execution from major cloud providers in the second half of 2026-any pullbacks or delays could disproportionately affect both AMD and Nvidia. Pay attention to margin trends for AMD as it scales production and competes on pricing. Also watch for developments in custom chip initiatives from Meta, Amazon, and Google, which could reduce total addressable market for both companies. Finally, monitor competitive dynamics as Intel and other players position their AI offerings, and whether customers continue diversifying chip suppliers or consolidate around fewer vendors.