The semiconductor industry is undergoing rapid transformation, driven by advances in artificial intelligence (AI), geopolitical shifts, and increased government investment in domestic production. As AI adoption accelerates, demand for high-performance chips is surging, while supply chain dynamics are being reshaped by evolving trade policies and national security concerns. At the same time, the indispensability of semiconductors in industries such as automotive, healthcare, and energy is driving the need for continuous innovation and strategic adjustments. Supply chain resilience and technological sovereignty are now top priorities for businesses and governments. Efforts are underway to diversify production and reduce dependence, but structural challenges remain. Export controls, restrictions on critical materials, and shifting trade alliances are redefining the semiconductor landscape, requiring companies to navigate increasing complexity while maintaining a competitive advantage.

Demand Analysis: Semiconductors Power Innovation and Everyday Life

Why is demand important?

Semiconductors are indispensable in today's world. Driven by rapid technological advancements and growing demand across various industries, the semiconductor market is experiencing strong and evolving demand. When demand outstrips supply, analyzing these dynamics can reveal multiple avenues for capitalizing on emerging opportunities.

As the backbone and enabler of data centers, artificial intelligence, autonomous vehicles, smartphones, and other emerging technology trends, the global semiconductor market is projected to grow from $627 billion in 2024 to $1.03 trillion in 2030, driven by broad advances across end markets.

Automotive sector

The automotive industry is undergoing profound transformation, driven by electrification, autonomous driving, and software-defined vehicles (SDVs). These trends are rapidly becoming industry standards, amplifying the role and value of semiconductors in modern vehicles.

With the electric vehicle (EV) market projected to capture a majority share by around 2030, demand for high-voltage power semiconductors, such as silicon carbide (SiC), will surge. Simultaneously, autonomous driving technology is likely to advance, with most vehicles reaching Level 2 and a growing number reaching Level 3. This development will drive up the semiconductor content per vehicle, from sensor and connectivity integrated circuits (ICs) to processing units. The rise of software-defined vehicles could shift vehicles toward regional architectures with centralized computing power, raising the performance requirements of automotive system-on-chips (SoCs).

The car of the future could become more than just a means of transportation—it could become a new kind of "home," a high-performance computer on wheels, seamlessly powered by semiconductors.

Electrification and connectivity

The automotive industry is currently undergoing a transformation characterized by electrification, autonomous driving, and connectivity. With the rapid expansion of the electric vehicle market (first in China, then in Europe, the United States, and other regions), original equipment manufacturers (OEMs) are increasingly investing in hybrid and electric vehicles. These vehicles are projected to account for approximately 50% of total vehicle sales by 2030.

The emergence of connected and autonomous vehicles is also shaping the future of the automotive market and driving its maturity. These trends, coupled with shifts in powertrain technology, could become the new standard in the automotive industry, elevating the role of semiconductors.

More electric cars? More electricity!

The rapid growth of electric vehicles, along with the integration of infotainment and autonomous driving, is increasing demand for power semiconductors. Power semiconductors are essential for managing and switching the electrical systems in modern vehicles.

As the automotive industry transitions from internal combustion engines (ICEs) to hybrid electric vehicles (HEVs) and pure electric vehicles (EVs), power semiconductors can account for more than 50% of total semiconductor costs.

More efficient? We need a more powerful chip!

With the shift to electrification, efficiently controlling power becomes more challenging as engine drive and control, as well as more features like autonomous driving and infotainment, rely on electricity. Since driving an electric vehicle involves repeatedly switching high-voltage power, demand is likely to surge for power semiconductors capable of efficiently handling higher power. If chips cannot withstand the high-voltage environment, serious operational failures such as fires could result.

This could lead to increased demand for new materials such as silicon carbide (SiC) and gallium nitride (GaN). Compared to silicon chips, these can withstand higher voltages and offer faster switching speeds, reducing power losses during switching. Consequently, automakers are using GaN in the speed-critical medium-voltage stage and SiC as a core component in the high-voltage, high-power path to achieve a balance between efficiency, weight, and overall system cost in electric vehicle powertrains.

Autonomous driving technology is categorized from Levels 0 to 5. Levels 0-1 provide driver assistance features such as collision avoidance and lane departure prevention. Level 2 enables partial autonomy, such as maintaining a safe distance from other vehicles on the road. Starting at Level 3, vehicles can operate without constant driver monitoring. Level 3 is suitable for highways, Level 4 extends to ordinary roads, and Level 5 eliminates the need for a driver entirely, with the driver acting solely as a "passenger." By 2030, the majority of new cars are likely to include Level 2 autonomous driving capabilities, and Level 3 autonomous driving capabilities may account for over 10% of all vehicle shipments.

The car's "eyes, brains, and muscles"

As the level of autonomous driving increases, vehicles require greater capabilities to collect and process data. This advancement increases the complexity of vehicle electronic architectures and drives up the cost of semiconductors for high-performance computing (HPC) and advanced driver assistance systems (ADAS). To enable autonomous driving, vehicles must be equipped with multiple sensors and connectivity chips to perceive real-time information, computing chips to process this data, and electronic control units (ECUs) to take action with minimal latency.

As a result, as vehicles become more automated, the number of chips installed and the average price per chip have increased significantly, driving growth in the automotive semiconductor market.

Software-defined cars change the way cars work

Have you ever discovered new features after updating your smartphone's software? Now imagine the same concept in cars. Software-defined vehicles (SDVs) enable new features through updates, without requiring hardware changes.

With the rise of SDCs, the industry is moving toward a zonal architecture, where different zones of the vehicle are managed by a central computer. This approach further simplifies wiring, reduces physical complexity, and significantly improves the stability of software updates.

This architectural shift is also reshaping the automotive semiconductor market. The number of electronic control units (ECUs), which previously handled a single function, is decreasing, while assuming more complex roles. The focus is shifting from individual ECUs to high-performance SoCs, AI accelerators, and high-speed memory chips. Connectivity chips for real-time data transmission and secure microcontroller units (MCUs) for software protection are also becoming increasingly important.

Automotive SoCs will integrate processing units such as graphics processing units (GPUs) and image signal processors (ISPs), but as computing demands surge and vehicle architectures become more zonal, the adoption of dedicated AI accelerators will also increase.

Semiconductor demand by application in 2030

The bubble chart shows the projected semiconductor demand for major applications in 2030. The orange dashed lines mark the mean on each axis and divide the applications into four quadrants.

Electrification and Automation

These two major trends are significantly impacting semiconductor demand. Electric powertrains primarily impact power semiconductors (such as insulated-gate bipolar transistors (IGBTs) and silicon carbide chips), while autonomous driving primarily impacts advanced driver assistance system (ECUs). Furthermore, demand for both electric and autonomous vehicles is growing simultaneously.

Furthermore, as autonomous driving technology and the software-defined car trend advance and expand, demand for related semiconductors such as automotive high-performance computing, sensors, and connectivity chips is expected to grow. Furthermore, there are expectations for upgrades in semiconductors related to body, infotainment, and passenger safety to improve the in-vehicle environment.

However, the market size of chassis and internal combustion engines is expected to decline gradually due to reduced technological innovation and stagnant market size.

Servers and Networks

Since the rise of generative AI (Gen AI) applications in 2022, the amount of data generated and processed has grown exponentially. From AI-driven automation and the proliferation of the Internet of Things to the increasing intelligence of automotive and industrial systems, data is no longer just an asset—it has become the cornerstone of modern digital infrastructure.

By 2030, the growing demand for computing power is expected to further drive the development of CPUs, GPUs, and AI accelerators, with high-bandwidth memory (HBM) continuing to be a key component supporting them. For servers in particular, major technology companies, including cloud service providers, have begun developing their own application-specific integrated circuits (ASICs) to reduce operating costs. Meanwhile, the expansion of 5G is likely to drive demand for computing power in network equipment and gallium nitride (GaN)-based radio frequency (RF) chips to enable ultra-high-speed, low-latency communications.

Servers and networks can become the backbone of the intelligent applications around us, powered by continued advances in semiconductors.

AI Data Centers and Next-Generation Connectivity

The rapid development of artificial intelligence and connectivity, along with customer adoption of advanced technologies, is driving increased demand for data centers and the servers they house to process this data. With data center investments by cloud service providers, colocation centers, and telecommunications companies, the global server market is projected to exceed $300 billion by 2030.

At the same time, demand for infrastructure supporting inter-server and inter-node connectivity is rising.

The demand for faster, broader, and more reliable connectivity is driving market growth for devices such as routers and modems, which form the backbone and infrastructure. This trend extends beyond single applications to encompass enterprise, public, and private networks.

Faster, bigger, smarter data centers

It's a cliché, but we now truly live in a world of data and connectivity. The number of connected devices, including cars, home appliances, smartphones, and personal computers, has never been greater. In addition to this growth in connected devices, consumers are demanding higher-quality entertainment, such as AR/VR/XR gaming and seamless video streaming. Furthermore, the launch of "ChatGPT" in November 2022 is prompting businesses and individuals to actively utilize AI services for a wide range of possible applications.

These applications generate and require massive amounts of data, and we're only just beginning to see the beginning. With the high demand for gaming and video streaming, and especially for AI, global data center power consumption is expected to more than double by 2030.

Data centers are critical resources for storing, processing, and managing data. They used to primarily serve businesses, but as demand has grown, they have become hyperscale, offering Internet as a Service. Now, driven by the demands of AI-specific applications, data centers are evolving once again into AI data centers, enhancing managers' ability to provide uninterrupted service to data center users.

The future of smart infrastructure

As AI applications drive an increase in the amount of data required for processing and the corresponding expansion of data centers, operational expenses for cooling and power have reached astronomical levels. Enterprises are now seeking more cost-effective ways to operate.

One approach to achieving this while simultaneously increasing AI performance is to use specialized data center chips. These chips are crucial for achieving high performance because they can handle the intensive computational demands of data processing more efficiently than general-purpose processors. To achieve the required performance levels, enterprises are turning to these specialized chips to meet their needs.

However, even standard chips designed specifically for data centers are designed for multiple customers and therefore include features that specific customers may not use. Consequently, large technology companies such as cloud service providers are developing AI accelerators specifically for their own data center applications.

By developing AI chips tailored to specific workloads, enterprises can achieve higher performance while reducing costs and power consumption. As data processing demands increase, the need for cost reductions also grows, and demand for AI accelerators is expected to rise.

As a result, the revenue share of AI accelerators in data center chips is likely to grow rapidly, reaching approximately 50% of the total data center chip market. In addition to AI accelerators, other data center-specific chips, such as data processing units (DPUs) and advanced memory chips like high-bandwidth memory (HBM), are also likely to see growth. HBM reduces data processing bottlenecks and supports high-performance GPUs, while DPUs offload network workloads from the CPU by handling data transfers. Sales of these data center-specific chips are likely to continue to grow as they become increasingly critical.

....

Source: Semiconductor Industry Observer

View more at EASELINK