Sturdy Momentum anticipated for Information Middle AI Chip Packaging in 2025-2030

TAIPEI, Dec. 9, 2025 /PRNewswire/ — 

Information Middle AI Chip Shipments Proceed to Develop

In line with the original document printed by means of DIGITIMES Asia, international information middle AI chip shipments are projected to develop from 30.5 million devices in 2024 to 53.4 million devices in 2030. This knowledge middle AI chip division comprises high-end and mid-range GPUs, application-specific AI chips (comparable to Google’s TPUs), AI server CPUs, and networking/interconnect-related chips (e.g., Transfer ASICs/rack-scale-up Interconnect Chips/DPUs & NICs).

In relation to expansion charge, the fastest-growing branchs by means of cargo quantity come with application-specific AI chips, mid-range GPUs that usefulness GDDR DRAM (comparable to Nvidia CPX GPUs), and interconnect-related chips, comparable to NVSwitch ICs. Amongst those 5 chip divisions, the cargo quantity expansion charges for high-end GPUs and AI server CPUs are somewhat decrease over the 2024–2030 duration, with a Compound Annual Expansion Charge (CAGR) of best about 10%. On the other hand, when bearing in mind the income from high-end packaging and end-market gross sales, the expansion charge can be a lot upper than the cargo expansion charge. The primary causes for this come with the larger content material price in step with chip and the adoption of complex packaging.

As an example, AMD not too long ago forecast that the worldwide server CPU income General Addressable Marketplace (TAM) CAGR will succeed in 18% between 2025 and 2030 (implying that the Moderate Promoting Value CAGR for server CPUs may exceed 7% because of the leap in core depend and the adoption of extra complex processes and packaging). For GPUs, in spite of somewhat average expansion in shipments, wafer foundry and packaging income CAGRs are each anticipated to exceed 40% in 2024-2030, pushed by means of the expanding selection of GPU and I/O dies in step with chip.

Information Middle AI Chip Packaging Income Extra Enthusiastic about GPUs

Not too long ago, application-specific AI chips like Google’s TPU and AWS’s Trainium have garnered vital consideration for his or her adapted power-efficiency traits, sparking hypothesis that they could exchange GPUs. In line with the original “Global Data Center AI Chip Packaging Market Forecast 2024-2030” document printed by means of DIGITIMES Asia, the total income proportion of knowledge middle AI chip packaging will stay GPU-centric right through the 2024–2030 forecast duration. It’s estimated that by means of 2030, packaging income for information middle GPUs will stay over 40% upper than that of application-specific AI chips.

Era AI server CPUs and AI networking-related chips have grand cargo volumes, the high-end packaging marketplace will nonetheless be overwhelmingly ruled by means of GPUs and application-specific AI chips.

Drivers and Inhibitors for Complicated Packaging of Information Middle AI Chips

DIGITIMES Asia forecasts that the worldwide marketplace dimension for complex packaging of knowledge middle AI chips will develop from US$5.6 billion in 2024 to US$53.1 billion in 2030, representing a CAGR of over 40%. This aligns intently with AMD’s fresh Monetary Analyst Year estimate that the worldwide Information Middle TAM CAGR will exceed 40% over the later 5 years.

Here’s a abstract of the primary drivers and inhibitors for the worldwide marketplace for complex packaging of knowledge middle AI chips:

Key Drivers:

• The AI current and the compute ability fingers race amongst tech giants.
• Complicated packaging using system-level scaling to increase Moore’s Legislation.
• Geopolitical festival between the U.S. and China, and the arise of Independent AI tasks.

Key Inhibitors:

• The re-assessment of go back on funding (ROI) relative to large computing ability investments.
• The have an effect on of less expensive and extra environment friendly unutilized applied sciences.

For extra main points on AI Chip, and the month outlook for main AI chip packaging applied sciences in information facilities, take a look at: AI Chip packaging https://www.digitimes.com/reports/ai/2025_ai_chip_packaging

SOURCE DIGITIMES ASIA