Here’s Why AMD Didn’t Rise After Q3 Earnings Results
With the reporting of its Fiscal Year (FY) 2025 results on the 3rd of February 2026, American chipmaker Advanced Micro Devices, Inc ( $Advanced Micro Devices(AMD)$ ) seems to be indicating that hasn’t made substantial headway in its battle for market share in the hottest sub-theme within AI: datacentres.
While AMD CEO Lisa Su stated during 3rd quarter (Q3) 2025 earnings call that its deal with OpenAI – a mirroring of the web of deals that Nvidia had made into the AI ecosystem – could generate more than $100 billion in revenue for the company, this doesn’t seem quite as evident at least for now.
Trend Analysis
In the final quarter of FY 2025, AMD largely maintains revenue trends estimated via the Q3 2025 results while delivering some pretty solid gains in diluted earnings per share.
The “top line” (i.e. revenue) results were achieved largely by the acceleration of sales to datacentres in Q4 2025. In Q3 2025, this segment was supposed to deliver a relatively modest 18% growth. By the end of FY 2025, this segment went on to deliver nearly twice the estimate – at 32% – for the entire year.
Around 1% of FY 2025’s revenue (and about 7% of Q4 2025’s “Data Center” segment revenue) was as a result of the Trump administration permitting the export of hitherto-restricted AMD Instinct MI308 GPUs to China in July 2025. AMD thus benefited from an approximate $360 million release of inventory and related charges to recognize revenue of approximately $390 million.
The release of inventory charges plus the modest improvement in operating income growth in the “Client and Gaming” segment, which went from a trend of 138% growth as of Q3 2025 to 141% for FY 2025, can be considered to have supported the outperformance in diluted EPS growth trend over the past quarter.
On the other hand, “embedded” revenue showed a slight acceleration in loss of growth from 2% in Q2 2025 trends to 3% by the end of FY 2025. While the “Data Center” segment did push its revenue growth trend up from the estimated 19% in Q3 2025 to a respectable 32%, the most resilient driver of both top and bottom line has been the “Client and Gaming” segment, i.e. its historical mainstay from before the AI Boom.
In segment-wise revenue share trends, a substantial uptick in sales to datacentre clients largely staved off most of the downtrend imputed in Q3 2025. However, it still closes FY 2025 with a 1% loss in revenue share relative to 2024.
Meanwhile, net revenue share of cost of sales as well as research and development (R&D) have been shrinking since 2023. Correlated with this is the year-on-year growth in the share of operating income as well as net income.
While the company states that it continues to explore and develop new AI-relevant products, it seems evident that there isn't a massive push unlike in 2021 and 2022.
New Partnerships and Market Conviction
Unlike Q3’s big “AI Deal” announcement with OpenAI, Q4 2025 brought a relatively modest but potentially transformative announcement: a partnership with India-headquartered multinational tech services giant Tata Consultancy Services (TCS). The partnership aims to assist enterprises globally to modernize hybrid cloud and edge environments, deploy AI-powered workplace solutions, and accelerate innovation across cloud-to-edge workloads. Also on the cards is co-developing GenAI frameworks for sectors such as life sciences, manufacturing, and financial services. The partnership expects to align industry expectations (via TCS’ deep involvement) relative to product design excellence (via AMD) and has no dollar value assigned.
The “Circular AI Deal Machine” employed by AMD’s rival Nvidia ( $NVIDIA(NVDA)$ ), while not necessarily likely to yield the massive dollar volumes promised and hyped throughout 2025, also locks in most major datacentre client demand in Nvidia’s favour. Datacentres are generally expected to be adhesive demand and GPU burnouts will likely require steady replacement of sold units to deliver consistent returns to the GPU provider. Meanwhile, AMD seems to be largely relegated to “off-datacentre” requirements among leading enterprises, which largely validates it’s “Client and Gaming” segment’s continued success.
With all factors considered, the muted response to AMD’s results in the post-trading and early pre-session trends on the 4th of January indicate that market conviction isn’t too enthusiastic likely because the most remunerative segment in the AI Hype – datacentres – isn’t likely to see a substantial shakeup in Nvidia’s market position.
Despite the solid gains in the net “bottom line”, some of the massive valuation imputed by its position within the AI Boom is likely to be tested. Rockiness in AMD’s trajectory can be expected through until the next quarter’s results.
Professional investors in Europe might consider the +3x Long AMD ETP (AMD3) and the -1x Short AMD ETP ( $LS -1X AMD(AMDS.UK)$ ) during bullish and bearish trends in AMD’s price. To potentially capitalize on major tech stocks seemingly driving the market currently, the 5x Long Magnificent 7 ETP (MAG7) and the -3x Short Magnificent 7 ETP ( $-3X SHORT MAGNIFICENT 7 ETP(MAGS.UK)$ ) are at hand.
Furthermore, the AMD Options ETP (YAMD) seeks to generate monthly income for investors by directly investing in AMD’s shares and selling put options on them. Also available is the Magnificent 7 Options ETP (MAGO), which invests in each Magnificent 7 constituent’s respective Options ETP in an equally-weighted manner.
***********
For broader articles that deep-dives into business and culture in Asia, visit asianomics.substack.com. Numerous new articles have been published that fully explain the rationale behind the commentary I’ve made in various European publications over the past few months. The latest one discusses the burgeoning number of military technology deals being enacted between Germany, France and India.
Comments