Geopolitical and Export Challenges Impacting AI Server Shipment Growth Forecasts for 2025
In a recent report, TrendForce has outlined the current landscape for AI server shipments for the year 2025, indicating a more tempered growth trajectory than previously anticipated. Despite major North American Cloud Service Providers (CSPs) driving the market forward, geopolitical issues and new export restrictions imposed by the US on China have led TrendForce to adjust its growth forecast to a 24.3% year-on-year increase for global AI server shipments.
The AI server market is heavily reliant on the demand from key players in North America, which remains stable thanks to continued investments from companies like Microsoft, Meta, and Google. These entities are not only focusing on AI server infrastructure but are also expanding their cloud capabilities, which in turn, bolsters hardware requirements. For instance, Microsoft has committed to utilizing NVIDIA's GPU-based solutions to support its AI infrastructure, although its own next-generation ASIC development progress has been slower than expected. The Maia chips from Microsoft are projected to commence scaling production in 2026.
Meta stands out as a significant player as well, reportedly ramping up its requirement for general-purpose servers due to the recent completion of new data centers. The company's enhanced focus on AI infrastructure has resulted in a notable uptick in shipments for its in-house ASIC chips, which are expected to double by 2026.
Google has also witnessed increased server demands spurred by sovereign cloud initiatives in Southeast Asia and additional data centers coming online. Having a relatively advanced adoption of self-developed chips, Google plans to mass deploy its AI inference-focused TPU v6e chips, a decision that has established them as a mainstream solution in the first half of 2025.
Amazon Web Services (AWS), on the other hand, continues to align its strategy around its Trainium v2 platform, while also initiating plans for multiple versions of Trainium v3 that are slated for mass production in 2026. Due to the expanding Trainium platform and an aggressive AI strategy, AWS is expected to double its ASIC shipments within the same calendar year, positioning itself as a leader among US CSPs.
Oracle, comparatively, has a more cautious approach, focusing predominantly on procuring AI servers and in-memory database server solutions. This year, Oracle intends to ramp up deployment of its AI server infrastructure along with the integration of its core cloud database services with AI applications.
As these major CSPs respond to shifting market dynamics, many original equipment manufacturers (OEMs) are reevaluating their strategies for the latter half of 2025 in light of changing international tariff policies. Despite the reassessment and potential headwinds posed by US export restrictions, the overall forecast for total AI and general-purpose server shipments remains a positive one, with expectation of approximately 5% year-on-year growth in total server shipments.
The upcoming months will be critical as major industry players adopt new strategies to navigate these geopolitical complexities, and the adjustments in shipment projections signal a market that is both resilient and adaptive in the face of global challenges. As TrendForce continues to monitor the situation, it remains to be seen how these developments will shape the future of the AI server market. Below, you can find the latest tech industry news, trends, and forecasts stemming from TrendForce's extensive research.
For more insights or inquiries on the shifting technology landscape, visit TrendForce's website or reach out to their sales department for detailed market data and reports. Established in Taipei in 2000, TrendForce has established itself as a significant analyst firm, gathering a reputed subscriber base and maintaining a dedication to providing accurate and comprehensive insights into the advancement of the technology industry.
The AI server market is heavily reliant on the demand from key players in North America, which remains stable thanks to continued investments from companies like Microsoft, Meta, and Google. These entities are not only focusing on AI server infrastructure but are also expanding their cloud capabilities, which in turn, bolsters hardware requirements. For instance, Microsoft has committed to utilizing NVIDIA's GPU-based solutions to support its AI infrastructure, although its own next-generation ASIC development progress has been slower than expected. The Maia chips from Microsoft are projected to commence scaling production in 2026.
Meta stands out as a significant player as well, reportedly ramping up its requirement for general-purpose servers due to the recent completion of new data centers. The company's enhanced focus on AI infrastructure has resulted in a notable uptick in shipments for its in-house ASIC chips, which are expected to double by 2026.
Google has also witnessed increased server demands spurred by sovereign cloud initiatives in Southeast Asia and additional data centers coming online. Having a relatively advanced adoption of self-developed chips, Google plans to mass deploy its AI inference-focused TPU v6e chips, a decision that has established them as a mainstream solution in the first half of 2025.
Amazon Web Services (AWS), on the other hand, continues to align its strategy around its Trainium v2 platform, while also initiating plans for multiple versions of Trainium v3 that are slated for mass production in 2026. Due to the expanding Trainium platform and an aggressive AI strategy, AWS is expected to double its ASIC shipments within the same calendar year, positioning itself as a leader among US CSPs.
Oracle, comparatively, has a more cautious approach, focusing predominantly on procuring AI servers and in-memory database server solutions. This year, Oracle intends to ramp up deployment of its AI server infrastructure along with the integration of its core cloud database services with AI applications.
As these major CSPs respond to shifting market dynamics, many original equipment manufacturers (OEMs) are reevaluating their strategies for the latter half of 2025 in light of changing international tariff policies. Despite the reassessment and potential headwinds posed by US export restrictions, the overall forecast for total AI and general-purpose server shipments remains a positive one, with expectation of approximately 5% year-on-year growth in total server shipments.
The upcoming months will be critical as major industry players adopt new strategies to navigate these geopolitical complexities, and the adjustments in shipment projections signal a market that is both resilient and adaptive in the face of global challenges. As TrendForce continues to monitor the situation, it remains to be seen how these developments will shape the future of the AI server market. Below, you can find the latest tech industry news, trends, and forecasts stemming from TrendForce's extensive research.
For more insights or inquiries on the shifting technology landscape, visit TrendForce's website or reach out to their sales department for detailed market data and reports. Established in Taipei in 2000, TrendForce has established itself as a significant analyst firm, gathering a reputed subscriber base and maintaining a dedication to providing accurate and comprehensive insights into the advancement of the technology industry.
Topics Business Technology）
【About Using Articles】
You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.
【About Links】
Links are free to use.