AMD reported second-quarter profits that jumped 163 percent year-over-year on revenues reaching $11.5 billion, yet the company's stock plummeted 10.5 percent following the announcement before settling 8.7 percent below its opening price, according to an article published by The Register on August 5, 2026. Despite securing multiple gigawatts worth of Helios commitments from major players including OpenAI, Anthropic, and Meta, Wall Street's response signals mounting concern over the chipmaker's exposure to an AI market that hasn't yet demonstrated sustainable profitability. The disconnect between AMD's robust financial performance and investor skepticism centers on the company's growing dependence on a small group of customers for its AI-focused products.
The company's datacenter and AI segment drove much of the quarter's growth, with CEO Lisa Su forecasting that datacenter segment revenue will more than double year-over-year in 2027. AMD's embedded division also posted solid results with revenues topping $977 million, representing a 19 percent increase compared to the same period last year. Client PC sales climbed 23 percent year-over-year to $3.1 billion during the quarter. However, the gaming division faced headwinds with revenue declining 31 percent to $779 million, pressured by the end of a console sales cycle that reduced semi-custom processor orders. Looking ahead, AMD issued Q3 guidance calling for revenues of $13 billion, plus or minus $300 million.
During Tuesday's earnings call, Su acknowledged the concentration risk while attempting to ease investor concerns, noting that while large frontier-model companies like OpenAI, Anthropic, and Meta will consume products through cloud service providers, "there are additional customers or lots of customers who are interested in Helios at, let's call it, a more regular scale than gigawatt scale." The Register observed that Su's statement revealed the underlying issue: AMD can sell substantial GPU volumes, but most go to a handful of buyers, while other interested parties don't purchase in bulk. Microsoft, another flagship customer for AMD's latest AI hardware, serves both OpenAI and Anthropic, while Meta is reportedly considering entering the GPU cloud business itself.
The company's optimism rests on recent product launches that position it as Nvidia's most credible challenger. At its Advancing AI event in San Francisco last month, AMD unveiled the Helios rack-scale platform packing 72 Instinct MI455X GPUs with 432 GB of HBM4 memory each, which on paper matches or exceeds Nvidia's Vera Rubin platform on most performance metrics. The company also introduced Venice Epycs offering up to 256 cores and 512 threads with support for 16 memory channels running at speeds up to 1.6 TB/s per socket, targeting demand for CPUs to power agentic AI sandboxes that Su anticipates will become the biggest growth driver for Epyc sales. However, as The Register notes, the CPU market is no longer the duopoly it once was, with AMD now facing traditional rival Intel plus new competitors including Nvidia, Arm, Qualcomm, AWS, Google, and Microsoft.
Su remains bullish despite these challenges, stating that datacenter AI business is positioned for significant growth in the second half of the year with growth accelerating in 2027, driven by increasing Helios and MI450-series deployments. The CEO expressed confidence that all of AMD's divisions—server CPU, datacenter AI, embedded, and PC—will benefit from AI tailwinds. Yet the company warned that the ongoing RAMpocalypse is likely to cut into PC sales over coming quarters, adding another variable to an already uncertain outlook. The chipmaker's forecast hinges on AI adoption continuing at current pace among customers that have yet to demonstrate they can operate profitably, a dependency that explains Wall Street's nervous reaction to otherwise impressive results. The market's hesitation suggests that even strong execution and competitive products may not be enough if the underlying AI spending boom proves unsustainable, leaving AMD caught between technological opportunity and financial prudence at a moment when both matter equally.