Majority of Chip-Dependent Firms Concerned About Semiconductor Supply in Two Years

The latest report from the Capgemini Research Institute on the semiconductor industry forecasts a significant increase in demand for advanced semiconductor solutions due to the growing adoption of artificial intelligence (AI) and generative AI (Gen AI). Although the industry is at the forefront of innovation, concerns regarding its ability to meet this demand are being influenced by geopolitical tensions, international trade limitations, and the ongoing movement towards sovereignty. The report indicates that the need for AI chips, custom silicon chips, and memory-heavy chips is projected to rise within the next year, highlighting the necessity for the semiconductor industry to seize new opportunities in design and innovative, sustainable manufacturing practices, coupled with investments in domestic sourcing and nearshoring to promote stability.

Gen AI and various emerging technologies including 5G, the Internet of Things (IoT), autonomous vehicles, augmented/virtual reality (AR/VR), and edge computing are creating a heightened need for more powerful, efficient, and tailored chips. Nearly 60% of semiconductor businesses report that Gen AI, 5G, or other advanced communication protocols are shaping their strategic direction.

While advancements in semiconductor technologies have fostered innovation in downstream sectors and contributed to the creation of smarter, more efficient products, less than 30% of downstream organizations believe the supply of chips is adequate.

“We find ourselves at a crucial juncture for the semiconductor sector. The rush for chips driven by Gen AI is mounting, and semiconductor companies are encountering increasing expectations from clients who demand more personalized, software-centric solutions,” stated Brett Bonthron, Global High-tech Industry Leader at Capgemini. “The industry should view this as an opportunity to boost production and embrace a ‘chip-to-industry’ strategy that supports a comprehensive, ‘software-first’ suite of capabilities. Investments in state-of-the-art fabrication techniques and design processes empowered by AI and Gen AI will be essential to address the specific needs of emerging applications. Additionally, it is vital for the industry to strengthen sustainable manufacturing practices and employ advanced security measures to protect intellectual property (IP).”

A significant rise in demand for AI chips and custom-designed chips

According to the report, while 39% of semiconductor organizations foresee Gen AI as a catalyst for custom chip demand over the next two years, a substantial majority of downstream organizations (81%) predict their demand will surge by 21% in just the next year.

As a result, downstream businesses and technology giants are increasingly considering in-house custom chip design, which enables them to create semiconductors tailored to their particular specifications. This approach reduces dependence on external suppliers, protects their intellectual property, and enhances speed, efficiency, and compatibility with various hardware and software.

Meanwhile, to cater to the demands of downstream organizations, the semiconductor industry persists in its pursuit of design and manufacturing innovations, pushing the limits of physics through advancements in chip architecture, extreme ultraviolet (EUV) lithography, finer process nodes, 3D packaging, and chiplet utilization. The report anticipates a roughly 10% increase in the industry’s R&D expenditure over the next two years.

Almost half of manufacturers are also leveraging AI and Machine Learning (ML) to refine their processes.

The integration of AI and Gen AI is fueling demand for more specialized, high-performance chips.

The increasing implementation of AI and generative AI is creating a stronger need for specialized neural processing units (NPUs) and high-performance graphics processing units (GPUs) that manage massive computations and substantial datasets efficiently. More than half of downstream industries (54%) believe that advancements in GPU computing and AI/machine learning enhancement will deliver the greatest value.

Chip sustainability, supply chain resilience, and security top the list of priorities for downstream organizations.

The report reveals that over half of downstream organizations are set to prioritize chip sustainability, supply chain resilience, and cybersecurity features within the next two years.

Only 40% of semiconductor organizations express confidence in the robustness of their supply chains. To mitigate risks tied to international logistics, the semiconductor industry is projecting an increase in domestic sourcing from the current 40% to 47% over the next two years. Additionally, a 4% rise in nearshoring is also anticipated. 74% of semiconductor companies expect to ramp up their investments in the U.S., compared to 59% who plan to increase investments in Europe.

Chip security remains a pressing concern in a highly complex and interdependent supply chain, with nearly 60% of semiconductor design organizations focusing on cryptographic protection.

Finally, with almost 60% of downstream organizations underscoring the importance of chip sustainability in their selection process, the industry is prioritizing key eco-friendly initiatives: energy conservation, implementation of water recycling and reuse systems, minimizing the use of toxic alternatives, and waste reduction.