SAN FRANCISCO: Advanced Micro Devices (AMD) has announced that its next-generation artificial intelligence (AI) server platform, Helios, has entered full production and will begin shipping to customers in the coming months, marking a major step in the company’s effort to challenge Nvidia’s dominance in the booming AI infrastructure market.
Speaking during AMD’s AI event in San Francisco on Thursday, Chief Executive Officer Lisa Su said customer demand for the new AI server platform has been exceptionally strong, with deliveries scheduled to begin toward the end of the third quarter of 2026.
“Helios is in full production, Su said. Customer demand for Helios is extremely strong.”
The Helios server combines AMD’s latest Instinct MI455X AI accelerator with its newly unveiled Venice central processing unit (CPU). Both chips are manufactured by Taiwan Semiconductor Manufacturing Company (TSMC) and are designed to power the next generation of AI training and inference workloads.
In a significant endorsement for AMD, OpenAI confirmed that it plans to begin deploying Helios server racks later this year.
OpenAI’s Vice President of Compute Strategy, Sachin Katti, revealed during the event that the ChatGPT developer expects to roll out the new infrastructure at large scale beginning late 2026, with broader expansion throughout 2027.
“We expect that we’ll be deploying Helios at massive scale starting towards the end of this year and then accelerating throughout 2027, Katti said.
He also disclosed that OpenAI intends to adopt AMD’s upcoming MI500 AI chips, highlighting the deepening partnership between the two companies.
AMD and OpenAI signed a multi-year collaboration agreement last year aimed at developing advanced AI hardware, strengthening AMD’s position as an alternative supplier in the rapidly expanding AI ecosystem.
Lisa Su projected enormous growth for the AI computing industry, estimating that the global computing market could reach $2 trillion by 2030, up sharply from an estimated $365 billion in 2025.
According to AMD’s projections:
- $1.4 trillion will come from AI accelerator chips.
- $220 billion will be generated by CPUs.
- The remaining market value will come from networking, storage and other computing technologies.
Su emphasized that no single company can satisfy the exploding demand for AI infrastructure.
“The only way you’re going to be able to service a market like that is for us to work together as an ecosystem,” she said.
AMD Expands AI Partnerships
AMD also unveiled several strategic partnerships designed to accelerate adoption of its AI hardware.
AI chipmaker Cerebras Systems announced that it will integrate AMD’s AI servers into its data centers, offering customers a combined AI computing platform.
Meanwhile, AI startup Anthropic recently signed an agreement to purchase up to two gigawatts of AMD’s Instinct MI450 AI chips beginning in the first half of 2027. The partnership also includes an investment of up to $5 billion in the Claude AI developer.
Anthropic co-founder Tom Brown demonstrated how Claude AI successfully configured AMD’s AI servers autonomously over a weekend.
“Anyone, human or AI, can now build real models on your platform,” Brown said.
AMD’s latest announcement comes as competition in the AI semiconductor market continues to intensify.
The company is targeting Nvidia’s dominant position in AI data-center chips, particularly in AI inference—the process of generating responses when users interact with AI models such as ChatGPT.
Nvidia this week also unveiled additional technical details of its upcoming Vera CPU and Rubin GPU platform, designed to maximize AI performance while reducing power consumption.
Although AMD shares slipped about 4% following the announcement, analysts attributed the decline largely to broader weakness across U.S. technology stocks rather than concerns over Helios.
Industry observers say AMD’s growing list of partnerships with OpenAI, Anthropic and cloud providers demonstrates that the company is becoming an increasingly important player in the global AI infrastructure race.
With AI spending expected to surge over the next decade, the launch of Helios represents one of AMD’s most significant attempts yet to narrow the gap with Nvidia and capture a larger share of the multitrillion-dollar AI computing market.
