downloadGroupGroupnoun_press release_995423_000000 copyGroupnoun_Feed_96767_000000Group 19noun_pictures_1817522_000000Member company iconResource item iconStore item iconGroup 19Group 19noun_Photo_2085192_000000 Copynoun_presentation_2096081_000000Group 19Group Copy 7noun_webinar_692730_000000Path
Skip to main content
Default Banner Image

AI

AI is redrawing the map of semiconductor innovation, challenging the industry to enable rapid advancements in high-performance computing. The next leap in computing will not come from a single breakthrough device but from how effectively the industry integrates logic, memory, photonics, packaging and software into complete systems, and ultimately, how quickly it can move those systems from lab to fab.Few organisations sit as close to this shift as imec. As one of the world’s leading research and innovation hubs in nanoelectronics and digital technologies, imec works with partners across the entire value chain, including materials and equipment suppliers, chip designers, foundries and system companies. Since April 2026, imec has been led by CEO Patrick Vandenameele, whose career spans senior industry roles at imec, research and deep-tech entrepreneurship. Ahead of his imec ITF Chip into the Future session on 10 November at SEMICON Europa 2026, taking place 10 to 13 November in Munich, SEMI Europe spoke with Vandenameele about the trends reshaping the industry, why collaboration matters more than ever, and the role Europe can play in global semiconductor competitiveness.What are the most significant technology and market trends shaping the semiconductor industry today?Vandenameele: The semiconductor industry is entering a new phase. For decades, progress was largely driven by scaling individual devices. Today, innovation increasingly happens at the system level, where advances in logic, memory, photonics, packaging, software and system architecture need to come together to deliver meaningful gains in performance, energy efficiency, and functionality. This shift is being accelerated by AI, which is creating unprecedented demand for computing power while raising the bar for sustainability and efficiency.At the same time, we are seeing growing opportunities at the intersection of different technology domains. Technologies such as chiplets, heterogeneous integration, silicon photonics, and advanced memory are becoming essential building blocks for future computing platforms. The challenge is no longer optimising individual components in isolation but orchestrating them into complete solutions that address real-world needs. To translate system-level breakthroughs into impact, the industry also needs a clear path from research and prototyping to scalable manufacturing, so that new technology platforms can be validated, industrialised and ultimately adopted in real-world products.Why is collaboration across the semiconductor ecosystem more important than ever?Vandenameele: The growing complexity of future systems is increasing the need for collaboration across the ecosystem. No single company, university, or country has all the expertise needed to address the challenges ahead. Future breakthroughs will depend on how effectively we combine knowledge and capabilities from across the value chain.This is particularly true in the age of AI. Developing the next generation of computing platforms requires close collaboration between material suppliers, equipment companies, chip designers, manufacturers, system companies, software developers, start-ups, and research organisations. CMOS 2.0 is a good example of this shift, requiring innovation and optimisation across the entire stack rather than within individual technology domains.The semiconductor industry has always been a highly interconnected ecosystem, but the need for collaboration is becoming even more important as technologies become increasingly heterogeneous and system-driven.What role does Europe play in strengthening global innovation and competitiveness in semiconductors?Vandenameele: Europe’s role is defined by its ability to connect complementary strengths across the semiconductor value chain. The region combines excellence in research, semiconductor equipment, manufacturing capabilities, chip design and system architecture, communications infrastructure, automotive technologies, and industrial systems. Together, these capabilities create an ecosystem where innovation can move efficiently from early-stage research to industrial deployment and societal impact. This must also include an environment in which deep-tech ventures can emerge from Europe’s research base, access advanced infrastructure and manufacturing pathways and scale into globally competitive companies.Europe’s opportunity is not to replicate what already exists elsewhere, but to build on its own strengths and connect them more effectively. That also means creating the conditions for a new generation of European innovators to emerge, scale and bring breakthrough technologies to the global market. Initiatives such as the NanoIC pilot line, which bridges semiconductor research and industrial deployment, demonstrate how Europe can bring together research organisations, industry and policymakers to accelerate innovation and strengthen competitiveness across the semiconductor ecosystem.What key message would you like attendees to take away from your imec ITF session at SEMICON Europa?Vandenameele: The future of computing will not be defined by a single breakthrough technology. It will be shaped by how effectively we integrate advances across the full stack, from devices and architectures to software and applications. The question is no longer how to build a better chip, but how to bring together multiple innovations into scalable solutions that address real-world needs, with a clear path from research and prototyping to manufacturing and industrial adoption.That shift requires a new level of collaboration across the ecosystem, and it is exactly what we want to explore at SEMICON Europa and ITF Chip into the Future session.Looking ahead, what opportunities and challenges do you see for the industry over the next five years?Vandenameele: The biggest opportunity lies in the transformative potential of AI and intelligent systems. We are only at the beginning of a technology cycle that will influence virtually every sector, from healthcare and mobility to telecommunications, manufacturing and scientific discovery. The semiconductor industry will play a central role in enabling these advances. Realising that potential will require breakthroughs that make AI systems more scalable and energy efficient. Technologies such as silicon photonics and CMOS 2.0 are helping address critical bottlenecks in performance, bandwidth and power consumption, paving the way for the next generation of computing platforms.At the same time, the challenges are significant. We need to continue improving performance while reducing energy consumption. We need to manage escalating system complexity. We need to ensure access to talent and strengthen pathways from research to industrial adoption. Start-ups can play a critical role in bringing disruptive semiconductor innovations to new markets, provided they have access to advanced infrastructure, deep-tech expertise, manufacturing capabilities and growth capital.Most importantly, we need to maintain the collaborative spirit that has driven innovation in our industry for decades.I am optimistic because the industry has repeatedly shown its ability to reinvent itself. The future will not be defined by a single technology, but by how effectively we bring together multiple innovations into scalable solutions that create value for industry and society.Closing ThoughtsVandenameele’s message is clear: the era of optimizing components in isolation is giving way to an era of system-level orchestration. Delivering on the promise of AI will depend on whether the industry can combine advances across the full stack, keep energy consumption in check, and build reliable pathways from research to industrial adoption. For Europe, the opportunity lies in connecting its complementary strengths to create the conditions for the next generation of deep-tech innovators to scale.These themes sit at the heart of SEMICON Europa 2026. Under the theme “Europe’s Next Era: Navigating the Perfect Storm,” the event brings together the full electronics manufacturing supply chain to explore how collaboration can turn technological complexity into competitive advantage.Join Patrick Vandenameele at the imec ITF Chip into the Future session at SEMICON Europa 2026, Tuesday, 10 November, Hall C1, Messe München.Patrick Vandenameele has served as Chief Executive Officer at imec since April 2026. He has played a pivotal part in shaping imec’s R D strategy and innovation agenda, ensuring the organisation’s research aligns with emerging market needs and technological developments. Patrick holds a PhD from KU Leuven and started his career at imec in 1996. In 2000, he transitioned to entrepreneurship, founding several deep-tech ventures, after which, he held senior positions at leading semiconductor companies, driving successful product launches and large-scale commercialisation. Upon returning to imec in 2017, he was instrumental in expanding imec’s venturing activities. Since 2022, he consecutively served as Chief Portfolio Officer and as Co-COO, spearheading strategic initiatives spanning business development, and portfolio management.Sitong He is communications manager at SEMI Europe.
Read More
Held at the Moscone Center in San Francisco on October 12, SEMI’s annual Market Symposium kicked off SEMICON West 2026, setting the tone of the show for the week. The Market Symposium was hosted by Clark Tseng, senior director of SEMI Market Intelligence. The agenda focused on the pressing issues facing the semiconductor industry, enabling the leading analysts and strategists to gain insight into the drivers influencing the semiconductor industry and to guide the day-to-day decisions needed to grow their businesses. While the 2025 symposium focused on the geopolitical landscape, trade policy, and AI-driven investment, the 2026 talks centered on the influence that AI is having on the semiconductor industry, and how it will drive the industry and economics past the $1.5 trillion mark to $3 trillion in revenue over the next decade.This year’s topics covered how semiconductors are building the critical infrastructure of AI, and what to expect in the near future, from chip types to power consumption to how AI is influencing the global engineering and supply chain—and, of course, what types of revenue the industry might expect as AI continues to scale. Speakers and talk titles this year includedBill J. Wiseman – McKinsey Company “Beyond Moore’s Law to Win the Race for AI Inference”Duncan Meldrum – Hilltop Economics “Macroeconomic Cross Currents: Are Semiconductors Vulnerable?”Kunjan Sobhani – Bloomberg Intelligence“AI Demand in the Next Semiconductor Cycle: GPU, ASICS and Beyond”Jean-Christophe Eloy – Yole Group “Powering the Next Phase of AI: The Role of Next-Generation Chips in Driving Industry Transformation”Syed Alam – McKinsey Company “Building the Backbone of the AI Revolution: The Semiconductor Technologies Powering Next-Generation Data Centers”Joseph C. Fitzgerald – Boston Consulting Group (BCG)“Scaling Artificial Intelligence, Reshaping Supply: How AI Is Transforming Semiconductor Demand, Engineering, and Global Footprints”Clark Tseng – SEMI“AI Demand Meets Manufacturing Reality: Capacity, Equipment, and Supply Chain Outlook”Economic InfluencesAs the talk titles suggest, the semiconductor market is being driven by the economics of AI. Clark Tseng and Duncan Meldrum pointed out that AI capital spending is driven by the top four hyperscaler companies. It is estimated that they will spend $750 billion in 2026 on data centers. The total market will be somewhat higher, as was highlighted by Kunjan Sobhani, suggesting the total capex for data centers would be $1.2 trillion in the US by 2027. According to Meldrum, economic growth is facing two competing forces moving in opposite directions. Inflation is having a negative impact, and AI is having a positive impact. Meldrum's research on AI shows that it is steadily being adopted, but companies are not seeing measurable results they can talk about. He sees a strong continuing AI investment in the U.S., with the risks being changes in policy or disruptions by society as the buildout continues. This suggests that AI will continue its robust growth for at least the near future, with Sobhani predicting a $2.8 trillion total available market (TAM) in 2030, which also includes behind-the-grid power requirements. The Dynamics and Drivers of AI The speakers agreed that the industry is moving into the agentic AI era, with inference becoming the dominant driver of AI growth and models for the foreseeable future. Bill Wiseman believes that as AI moves into the agentic era, it will further reshape AI demand. Wiseman’s colleague Syed Alam pointed out that inference will account for 42% of AI demand in 2030. Strong token demand will drive chips, data centers, and power exponentially over the next several years, according to Wiseman, Alam, Sobhani, and Joseph Fitzgerald.This will place a tremendous strain on the industry as it struggles to keep pace with the demand. Fitzgerald pointed out that AI demand is exceeding supply by 60%. Wiseman stated that AI is growing exponentially while supply, or current processes, are growing linearly, and each of the speakers gave thoughts on ways to deal with the bottlenecks and cost. There is a consensus that the cost per token per joule needs to come down considerably. To achieve lower token cost, the speakers agreed that the industry needs to think differently. Alam suggested using six interconnecting dimensions that will drive a systems-level approach for the future of AI.By addressing each of these six dimensions as an interconnected system, the industry can address and reach solutions for the pressing issues challenging AI growth. Wiseman had slightly different ideas regarding the approach to meeting demand. Using a Peter Drucker quote—"The greatest danger in times of turbulence is not the turbulence; it is to act with yesterday’s logic”—Wiseman then summarized: AI is growing exponentially, and the supply chain is linear. The industry needs visionary thinking that is not business as usual but instead challenges the fundamentals of how integrated circuits currently work together to achieve the needed growth. Practical Use of AI and Supply Chain ShiftsFitzgerald dealt with some of the more practical aspects of how AI is transforming design. AI has demonstrated gains particularly in product design and design verification. The process moves from sequential to continuous, and engineers are used for final validation, where judgment is required. It is crucial to explore model and agent costs, as they are coming down and differ by model, and the cost savings can be significant.Fitzgerald also shared information on the supply chain, which is shifting due to tariffs, geopolitical concerns, and the host-nation environment. He concluded by noting that the supply chain will not reset; this means that supply chain leaders should continue to optimize for regional networks through 2030, and likely beyond. The integrated global supply chain will not return. Regional hubs are replacing it, and winners will pick a clear position, commit early, and execute with disciplineNow, down to why the symposium is a must-attend event: the industry outlook. The AI and Semiconductor OutlookIt's no secret AI is driving the semiconductor industry to new heights. How the industry thinks about forecasting is changing to gigawatts (GW) and tokens per joule. The rapid increase in AI activity and models, along with the rising average selling prices (ASP) of DRAM and accelerators, has made forecasting a moving target, according to Jean-Christophe Eloy. In 2031, the AI market is estimated to be demanding between 245-300 GW of power, according to Fitzgerald and Alam. The high side will see an accelerated market; the low side of 245 GW will be constrained by power and memory supply. AI inference is expected to be the key growth driver, with inference accounting for 70% of AI in 2025, growing to 90% of AI compute in 2040, according to Wiesman. This will drive a $2.8 trillion data center TAM, which does not include China, according to Sobhani. He also reported that the accelerator market, which includes GPUs and high-bandwidth memory (HBM), will grow from $400 billion in 2026 to $1.4 trillion in 2030. GPUs and ASICs will expand together in this market. Eloy is in close agreement, citing $464 billion in data center processor revenue in 2026, which includes GPUs, CPUs, networking, and ASICs. This expands to $1,631 billion in 2030.Total semiconductor chip revenue in 2026 is forecast to be $1.5 trillion by SEMI and $1.7 trillion by Yole. As mentioned earlier, these are revisions from Q2, and the increases are driven by the rapidly growing ASP of DRAM, as well as accelerators. The market for 2030 is forecast to be $2.3 trillion by SEMI; Yole’s prediction is $3.5 trillion. While the bulk of the growth is attributed to accelerators and HBM, according to Yole, power semiconductors are expected to see significant growth—from $15,000 per AI rack to $100,000 by 2030. To drive this growth in semiconductors, SEMI forecasts that 300mm equipment spending will grow from $149 billion in 2026 to $227 billion in 2029, thanks to double-digit growth each year over the time period. The global wafer fab equipment market will rise from $117 billion in 2025 to $220 billion in 2028. Foundry/logic remains the largest segment, increasing from $65.6 billion to $116 billion as AI/HPC and advanced-node ramps continue. DRAM investment expands strongly, reaching $63 billion by 2028, supported by HBM and advanced-node migration and capacity expansion. The growth in AI is also driving packaging and assembly to new heights. Growing from $6.1 billion to 10 billion in $2028. Materials are expected to grow in 2026 and 2027 by 12.85 and 11.4%, reaching $92 billion in 2028. Wrap-upThe SEMI Market Symposium laid out the industry journey that AI is piloting. While there are discussions of a bubble, the information shared suggests strong growth for the semiconductor industry through 2030, with semiconductor revenue potentially surpassing the $3 trillion mark by then. Equipment and material revenues will keep pace, albeit at a slightly slower growth rate due to market dynamics. SEMI would like to thank Market Symposium speakers and attendees for their participation at the SEMICON West 2026. For additional market insights, explore the latest SEMI Market Intelligence reports, covering historical reporting, actionable foresight into emerging trends, and technology investments to make confident, forward-looking decisions across the semiconductor and microelectronics value chain.Please join us in Phoenix on March 29 for the SEMI Market Symposium at SEMICON West 2027. Dean Freeman is Technology Advisor at Bodewell Group.
Read More
The headlines are dominated by the unprecedented growth in data centers and their rising energy consumption, along with the skyrocketing user cost of AI for inference at scale. Perhaps less obvious is the direct connection between the two – reducing energy consumption reduces AI cost. This is because both require smart use of AI system capability at maximum efficiency, which is often not the case today.Energy efficiency must be a strategic priority for all companies using AI inference, not just for data center builders. Smarter AI usage with lower energy consumption and less cost will help enhance the financial bottom line. Leadership in energy efficiency innovation will also generate a topline reward for companies expanding into AI’s next frontier, physical AI – when AI ventures off the computer screen into the real world. A robot, a drone, or a wearable typically run on a limited battery, so energy efficiency is a must-have, not optional!Each of these challenges can turn into opportunities.EnergyAI training compute is growing at an estimated 4–5x per year, with the latest large language models (LLMs) reportedly using 5-50 trillion parameters. The rapid growth of AI inference at scale amplifies this massively and generates enormous energy demand. In the US, data center energy consumption has tripled over the past decade and by most projections, it will triple again in just 5 years! The International Energy Agency (IEA) estimates that globally, data centers consumed ~460 terawatt-hours (TWH) of electricity in 2025, which will more than double to 945 TWH by 2030. For perspective, this is more power than the entire country of Japan uses today!The global energy infrastructure was not designed to absorb this level of demand. Communities across the world are pushing back against this strain on their power grid and water supply, as prices rise and shortages occur. Nuclear energy may help augment the traditional grid, but it is a distant hope, at least a decade away. This challenge cannot be ignored and could become a major roadblock for the future of AI. It is no surprise that energy is now the strategic currency for AI globally, with AI business transactions now being announced in Gigawatts, rather than Megaflops.CostInvestment in AI system infrastructure is rising at a dizzying pace: top players invested ~$100B for data centers in 2020, and this will increase TENFOLD to ~$1 trillion in 2026! Broader global investments are even higher. But there is no free lunch: everyone must pay! AI users are now feeling the pinch as they use more AI for inference, and even big players have reportedly blown through their annual AI budgets in just months. To make matters worse, only a small fraction of AI applications currently generate financial value and provide the expected return-on-investment (ROI). In part, this is because basic AI capabilities are becoming table stakes: companies invest in expensive AI tools that may enhance a product or service, but competitors do the same; so there may be no price premium to be had. AI tokens have become the basic unit for quantifying workloads and cost, as AI providers typically bill usage in millions of input and output tokens. Daily token consumption in the US is projected to increase TENFOLD from today’s ~200 trillion units per day (TUs/day) to ~2250 TUs/day by 2030. Token prices are falling, but not fast enough to keep up with the rising token usage; so AI costs will continue to rise sharply on the current trajectory. Little wonder that CFOs and finance departments are sounding alarm bells and beginning to ration AI budgets.Physical AIAI has grown from data analytics to generative AI and is now crossing another frontier by stepping off the computer screen into the physical world. “Physical AI” – intelligent systems like robots, drones, or medical wearables – require fusion of software intelligence with mechanical, electrical, and optical systems like sensors, actuators, cameras, radar, LIDAR etc. Physical AI systems must be autonomous because they cannot rely on data center communication, which adds latency and security risks. So, they must possess “edge intelligence” for real‑time decision‑making to sense, think, and act continuously in the physical world. Each of these functions consumes energy, and an autonomous unit usually operates on a limited battery. Without breakthroughs in energy efficiency, these physical AI systems would either become tethered to power sources or operate with short duty cycles of limited usefulness. The Path ForwardInnovation is the magic wand that can turn these challenges into opportunities. The technology industry excels at innovation, but the spotlight needs to shift from “ever-larger” to “smarter and efficient” systems. Further, point solutions in silos are no longer sufficient – radical efficiency improvement requires system-level optimization of the entire AI stack. Join SEMI’s Smart Data-AI Initiative with our Alliance Partner, City of San Jose, for an insightful workshop on September 9 where we will explore how to bend the curve for energy, cost and performance for future AI computing. We unite leading industry experts across the whole AI ecosystem to see the big picture for future materials, devices, and systems; with deep dives on photonics, hardware-software co-optimization, and chip-to-grid enhancements. This is not yet another AI conference – it is a workshop to explore practical solutions and amplify your business strategy by connecting the dots across the AI stack. You’ll learn about cutting-edge innovations, network with experts, and explore meaningful collaborations.Frequently Asked QuestionsWhy is energy efficiency becoming a strategic priority for AI computing? AI training and inference are driving massive increases in data center energy use, putting pressure on power grids, water resources, and operating budgets. Mitigating the skyrocketing cost for AI users is also a growing strategic priority. Reducing energy consumption directly reduces cost of AI for users, since it implies that AI system capability is being used at maximum efficiency.How are AI tokens connected to rising AI costs?AI providers typically bill usage based on input and output tokens, making tokens a key measure of workload and cost. As daily token consumption grows, total AI spending can rise even if token prices fall, prompting companies to look for more efficient ways to use AI.Why does physical AI make energy efficiency even more important?Physical AI systems such as robots, drones and medical wearables must sense, think, and act in real time, often while running on limited battery power. Without major gains in energy efficiency, these systems could face shorter operating times, reduced usefulness, or dependence on external power sources.How can companies and industry stakeholders help drive innovation in smarter, more efficient systems?Companies and industry stakeholders can engage with SEMI’s Smart Data-AI Initiative by joining workshops, connecting with experts across the AI ecosystem, and participating in discussions on materials, devices, systems, photonics, hardware-software co-optimization, and chip-to-grid innovations. These forums are design to help organizations identify practical solutions, explore meaningful collaborations, and shape a more energy-efficient future for AI computing.Dr. Pushkar P. Apte is Global Lead for Smart Data-AI Initiative Strategic Technology Advisor at SEMI.
Read More
Artificial intelligence (AI) is scaling at a pace that is reshaping semiconductor roadmaps, data center design, and long-term infrastructure strategy. AI promises many economic and social benefits; but the growth comes with an escalating demand for power, and energy has emerged as a major challenge.SEMI, as the global semiconductor and electronics association connecting over 4,000 companies, continues to unite the entire ecosystem to “bend the curve” – to maximize AI performance while minimizing power consumption. In a series of successful, sold-out workshops that the SEMI Smart Data-AI Initiative held on this topic, a resonant theme has emerged: sustaining AI progress requires energy-efficient computing with holistic co-design and co-optimization across materials, devices, systems, data transmission, data centers, emerging architectures and software. While this dialog is an important starting point, the ultimate goal is to drive concrete action through collaborative innovation.The AI Energy ChallengeAI training compute for frontier models is growing at an estimated 4–5x per year, driving unprecedented demand for hardware capability and infrastructure capacity. That trajectory has resulted in a global “data center gold rush” and is testing energy availability limits. As model sizes scale exponentially, so too does the energy required to train and deploy them; and power consumption has become a significant limiter to performance gains. Further, this increases heat dissipation, and requires innovations like direct liquid cooling.Modern AI and high-performance computing systems now operate at levels comparable to small cities, with tens of megawatts per installation and a trajectory toward gigawatt-scale data center campuses. Grid capacity—both in the U.S. and globally—may be challenged to keep pace with projected demand. Thus, AI infrastructure is no longer just a technical challenge, but it is an energy, systems, and policy challenge.System-Technology Co-OptimizationContinuous advances in chip and inference efficiency have delivered orders-of-magnitude improvements over many decades. These gains must now be expanded by holistic co-optimization of the entire compute system from silicon technologies to data center to the grid.For example, processors can be made more efficient by customizing them for specific workloads. However, only part of total data center power is consumed by the processor itself. A significant portion is used by data movement, power conversion and cooling. The energy required to move data increases dramatically with distance. Moving bits across packages, boards, and networks can consume far more energy than the compute operations themselves. This makes locality a critical design principle. The opportunity—and necessity—therefore lies in cross-layer optimization: efficient compute, efficient communication, and intelligent power management across the entire system. Not surprisingly, advanced packaging and integration are becoming central to performance. These technologies can enable architectures that tightly couple compute, memory, and I/O—using 2.5D and 3D integration techniques—reducing energy per bit and increasing bandwidth. Photonic interconnects and low-power materials can further lower the cost of processing and moving data.The bottom line is that incremental chip-level gains alone will not be sufficient and energy optimization cannot be siloed—system-technology co-optimization is needed.Hardware-Software Co-optimizationKeeping data as localized as possible depends as much on software algorithms as it does on hardware architectures. The challenge is that the development cycles are mismatched: new software models can be developed in months, while designing and fabricating new hardware can take years. While this cycle mismatch is fundamental, closer coordination between hardware and software developers can significantly improve efficiency. For example, offloading selected functions in the algorithm, including distributed DPUs, and reducing the level of data precision can reduce energy use. Partitioning workloads logically across the hardware/software stack between cloud services and compute-on-edge can also reduce energy appreciably. Further, risk mitigation techniques—for example, building in strategic redundancy—can make future designs more resilient to shifts in software algorithms and models.Diverse Computing ModalitiesWhile AI dominates current infrastructure investment, the future of computing will likely include multiple, diverse computational modalities such as quantum, neuromorphic, photonic and analog computing.Different computational paradigms will be applied where they are most effective. For example, quantum computing is likely to complement—not replace—classical systems; especially for specific classes of problems where it offers exponential advantages. However, progress in quantum computing is tightly coupled to advances in semiconductor infrastructure. Error correction, orchestration, and hybrid algorithms all depend on high-performance classical systems operating with low latency alongside quantum processors. While there is no single silver bullet, system-level design can ensure that multiple computing modalities work together within unified workflows spanning edge, cloud, and exascale environments.Why It Matters What to WatchEnergy will now be a key constraint for AI performance and infrastructure expansion.The evolution of gigawatt-scale AI campuses and their interaction with public energy grids will accelerate – or slow down – AI growth.Data movement, memory bandwidth, interconnect efficiency, advanced packaging and heterogeneous integration will be strategic levers. Enhanced system-technology co-optimization and integration of advanced technologies like 3D ICs and photonics will be critical.Co-optimization across hardware, software, and systems will be required.Future architectures will blend classical and emerging compute modalities like quantum, photonic and neuromorphic.In conclusion, AI has become a defining global force with much promise, but its trajectory will be shaped by technology, energy and infrastructure economics working together. This is a formidable challenge because it requires many diverse players with divergent priorities to collaborate effectively.We invite you to join the SEMI Smart Data-AI initiative to collaboratively address this challenge and help realize AI’s full potential sustainably. Our next workshop in this series will be on September 9 in Silicon Valley – please join us for this exciting event.SourcesSEMI Smart Data-AI Initiative – Future of ComputingEnergy-Efficient Computing for AI and Beyond, SEMICON West, October 2025Sustainable AI Systems, SEMI HQ, March 2026About the AuthorsDr. Pushkar P. Apte is the Strategic Technology Advisor for SEMI Global Lead for the Smart Data-AI Initiative Dr. Melissa Grupen-Shemansky is Senior VP and CTO of SEMI
Read More
With edge AI emerging as a clear driver of smart manufacturing, SEMI hosted a two-day workshop detailing the future of this technology. The workshop, Smarter Sensors, Smarter Fabs: AI at the Edge in Semiconductor Manufacturing, was held in-person from March 18-19 in Milpitas, California. It convened industry professionals to explore how AI-driven sensors and edge intelligence are fostering scalable and resilient solutions for the next generation of semiconductor manufacturing. The workshop took place across four sessions, with each highlighting a unique edge-AI implementation area – including process control, yield enhancement, tool coordination, and predictive maintenance – and featured keynotes from leaders at Lam Research and KUKA. Didn’t get a chance to attend in person? View this workshop on demand. Session 1 - Smart Sensors and Edge Intelligence for Advanced Process Control​The semiconductor industry has always been defined by precision. However, as device architectures shrink to angstrom-scale dimensions, and as wafers become thinner and more fragile, traditional process control tools are reaching their limits. Sampled, low-frequency, univariate monitoring systems were built for an era where deviations were visible, failures were catchable later, and a handful of sensors per tool were enough to keep yield in check. Session 1 explored the latest sensor technologies, discussing how data collection at the point of production, with AI embedded directly into the tool, is becoming paramount for success. Advanced in-situ sensors were brought up as an example of this in practice. Although these sensors are generating richer signals than in the past, reaching lower latency requires AI models deployed at the edge.In addition, AI is extending into the physical world through robots that can handle various tasks autonomously. These robots are enabled by digital twins that provide simulation environments for training and validation before they ever see the fab floor. The common thread across Session 1 was the growing need for data and knowledge integration in fabs. Smart sensors must be built into AI systems, and those systems must be scalable across tools without sacrificing speed or reliability. Finally, the insights they generate must flow back into maintenance optimization and equipment health monitoring to promote a continuous cycle of learning. Session 2 - Yield Enhancement Through Edge-Driven Defect Detection and Classification​ Session 2 focused on how edge AI models, process sensors, and image data can identify yield-impacting defects earlier in the manufacturing process. As semiconductor devices lean into 3D architectures, the complexity and volume of data have outpaced the capabilities of traditional monitoring tools. Today's fabs are required to evaluate terabytes of inspection images per hour, as well as tool sensor traces that require analysis across dozens of parameters simultaneously. Each speaker approached this challenge from a different angle, yet the solutions fit together into a coherent architecture. One introduced Gaussian Process Regression, a model for assessing both predictions and uncertainties, as a statistically rigorous, data-efficient method for learning "golden trajectory" baselines for tool sensor signals. This generates actionable scores and maintenance guidance beyond standard anomaly alerts. Another speaker demonstrated the ability of deep learning models to triage multi-gigabit-per-second image streams in milliseconds. AI-based defect classification was shown to compress root cause analysis timelines from days to hours, with demonstrated gains of a 0.3% die yield recovery and 0.5–1% yield exposure prevention. Predictive metrology for RF filter frequency also assessed device performance using upstream process data, with less than 0.02% error.Lastly, a software-defined automation framework built on open standards and vendor-neutral architecture demonstrated effective workload consolidation onto a single edge platform. It was shown to be scalable across fabs without replacing legacy infrastructure.These presentations stressed the importance of measurement and action in real-time at the tool level. Gathering information as early as possible, using AI to triage and classify, and feeding insights back into process control and maintenance workflows, allows for a continuous cycle of improvement.Session 3 - Autonomous Work in Process Movement: Robots, Sensors, and Edge AI Coordination​The "lights out" factory is shifting from an aspiration to a concrete, engineering roadmap. To fully realize this, each presentation in Session 3 highlighted the importance of supplementing human-dependent workflows with AI systems that can act in real-time. This shift will require a mix of deep reinforcement learning and AI-based perception approaches. Currently, deep reinforcement learning is training agents to discover new routing strategies that optimize yield, equipment effectiveness, cycle time, and queue-time compliance – including joint front and back-end-of-line coordination for advanced packaging. AI-based perception is also on its way to replacing manual, pre-shipment inspection checklists, demonstrating inspection time reduction by as much as 78%. To enable these improvements, presenters suggested private 5G as the foundational connectivity infrastructure. Currently, private 5G is helping eliminate dead zones and bandwidth issues that are preventing real-time machine data and connected robotics from reaching their full potential.Based on these presentations, the prevailing formula is to integrate intelligence at every level. This includes precise in-situ sensing to eliminate manual setup and measurement, edge AI models that act on data immediately, platforms that coordinate across tools without humans, and lastly, a reliable connectivity infrastructure.Session 4 - Predictive Maintenance at the Edge: From Vibration to VisionSemiconductor fabs have long operated in a state of crisis management. Fab managers spend between 40% and 70% of their time firefighting unexpected equipment failures, rather than executing planned maintenance strategies. Unplanned downtime in semiconductor manufacturing can cost up to $1 million per hour, yet the maintenance industry has been slow to move beyond reactive repairs. Fab managers need faster ways to determine issues and act on that knowledge before wafers are lost. Session 4 outlined a framework for how this transformation will happen. At the foundation, smarter sensors (vibration, acoustic, thermal, spectral, and vision) are generating the high-fidelity, multi-modal data streams that make predictive models possible. In addition, "ultra edge" AI accelerators are enabling machine learning inference to happen directly inside MEMS sensors and on-device hardware without cloud dependency. Fabs require low-latency, data-sovereign, real-time decisions that the cloud is unable to support, and the path forward requires an integrated chain of sensing, edge inference, health scoring, and maintenance scheduling. This session also made the case that irrelevant correlations and confounding variables make purely statistical AI unreliable for root cause analysis, and that causal AI models are required to give fabs actionable information. It concluded that cybersecurity concerns, soaring cloud infrastructure costs (datacenter GPU prices reaching $25,000–$50,000 each in 2025–2026), and latency requirements have made distributed, machine-local intelligence the only viable path to achieving autonomous fabs. SummaryThis workshop highlighted how edge AI, smart sensors, and advanced connectivity are transforming semiconductor manufacturing by enabling real-time process control, faster defect detection, and more autonomous operations. Across sessions, experts emphasized that integrating AI directly at the source of data is essential for improving yield, reducing downtime, and building scalable, resilient “smart fabs.”Learn more by registering for this workshop on demand, or view the recap videos on LinkedIn. Day 1 recap Day 2 recap The SEMI Manufacturing Coalitions include Smart Manufacturing, Fab Owners Alliance (FOA), MEMS and Sensors Industry Group (MSIG), Advanced Packaging Heterogeneous Integration (APHI) and Semiconductor Components, Instruments, and Subsystems (SCIS).Anshu Bahadur leads the Smart Manufacturing Initiative, Karim Somani leads the Fab Owners Alliance (FOA), and Paul Carey leads the MEMS and Sensors Industry Group (MSIG), all of which are part of the Technology Coalitions at SEMI.
Read More
The artificial intelligence revolution has a dirty little secret: for all its "brain" power, AI and autonomy are hopeless without the hardware that connects them to the physical world. Today’s semiconductor-related blockbuster tech topics like autonomous humanoid robots, Edge AI, self-driving cars, and lights-out fabs all depend on a variety of sensing modalities and enabling technologies (from MEMS to photonic and others). The 2026 MEMS Sensors Executive Congress (MSEC) gathered industry leaders to discuss how sensors continue to evolve from simple sensing components to the essential “eyes and ears” of a global, AI-driven transformation.The Sensor Industry is ShiftingMarcellino Gemelli of Bosch Sensortec took a retrospective approach during the Leadership Roundtable starting with the statement “to understand where we are going, we have to look at where we’ve been.” While logic and memory chips scaled rapidly, sensors faced a different reality:Commoditization: Rapid price erosion in high-volume markets like mobile.The "One Process, One Product" Curse: Unlike standard CMOS, every new MEMS device historically required a unique manufacturing flow which results in high development costs.Hard to fill MEMS Fabs: Geometries are shrinking resulting in more devices per waferPackaging: A challenge because it directly impacts device performanceSensor Fusion: Integrating sensor components with ASICs and MCUs to create smart sensorsAccording to Maximize Market Research, the global sensor market is predicted to have an 8.7% CAGR from 2024 through 2030. There is a fundamental realization about sensors: the next stage of autonomous manufacturing and intelligent systems cannot exist without high-fidelity, real-time sensor data from the edge. This was a common theme throughout MSEC. Data presented at MSEC by Pierre-Marie Visse of Yole Group shows the global MEMS market is projected to grow more slowly with a 3.7% CAGR over the same time frame, with higher growth predicted for automotive, industrial, and medical applications.The Leadership Roundtable, featuring executives from Bosch, Infineon, STMicroelectronics, and Rogue Valley Microdevices, highlighted the strategic roadmaps that will define the next decade, echoed by others during other technical presentations:The Edge of Perception: AI is pushing sensing technologies to process data within the sensor itself and not in the cloud, reducing latency and power consumption while improving privacy.Autonomous Manufacturing: Leaders like John Behnke (INFICON) and Edvard Kälvesten (Silex) mapped out the path toward “Autonomous Fabs,” where sensors allow tools to communicate and self-optimize with minimal human intervention.Emerging Modalities: Beyond traditional motion and pressure sensing, MSEC spotlighted the rise of Quantum sensors for resilient navigation and Photonics combined with MEMS for ultra-precise inertial sensing.From “Parts” to “Interfaces”A recurring theme throughout the congress was the death of the "sensor as a part" mentality. In his keynote, Kurt Busch (Syntiant) argued that sensors plus AI models are becoming the default interface layer for products.“The next-generation interface is not a screen. It is the physical world, captured by sensors, interpreted by models, and delivered through natural interactions,” said Busch.This shift is visible in the rapid adoption of Edge AI integration, the development of humanoid robots, autonomous drones and vehicles, and AI enabled smart glasses—rewriting what the human machine interface looks like.Conclusion: Sensorizing the FutureWe are no longer just building devices; we are building an “industrial AI operating system” that connects the digital and physical world. This all starts with sensors. By 2030, the most valuable AI won't just be the one with the biggest brain, it will be the one with the best senses.If you are a leader in the field of MEMS Sensors, let your voice move the industry needle by becoming a SEMI and MEMS Sensors Industry Group (MSIG) member company and getting involved with MSIG. MSIG will be hosting the MEMS Sensors Technical Congress (MSTC) on September 16-17 at SEMI HQ in Milpitas, CA. Engineers and technical executives will dive deep into new technology and processes that advance the sensor industry. To learn more or to be a part of the fascinating world of MEMS Sensors visit the SEMI MSIG website.Paul Carey is Director, MSIG at SEMI. Rafael Tudela is Senior Technical Marketing Manager at SEMI.
Read More
The rapid growth of AI has created a surge in the global energy consumption at a rate never seen before. Today, data centers account for approximately 415 terawatt-hours (TWh) of electricity globally. To put this into perspective, the annual energy consumption of the United Kingdom in 2023 measured at 309 TWh. The International Energy Agency (IEA) projects data centers’ energy consumption will more than double to nearly 945 TWh by 2030 [1]. A single generative AI query can consume up to ten times the power of a traditional search [1]. Meanwhile, data center energy usage in the U.S. is projected to leap from 4.4% to as much as 12% of the national grid by 2028 [2]. This creates a stark reality for the semiconductor industry. Traditional monolithic scaling has hit its physical and economic limits, leaving advanced packaging and heterogeneous integration to define the industry’s trajectory [3].To meet these escalating compute demands, the industry is rapidly shifting toward multi-die architectures, chiplets, and 3D stacking to decrease the amount of energy needed for advanced computing. This transition is fueling explosive growth in the advanced packaging market, which the Yole Group projects will reach $79.4 billion by 2030 [4]. However, stacking chiplets to bypass Moore’s Law exposes massive systemic bottlenecks. Engineers are now fighting interconnect parasitics, navigating complex power delivery architectures, and battling extreme thermal density.In a 3D-stacked architecture, pulling heat away from vertically integrated dies is one of the most pressing engineering challenges of our time. As compute density rises, issues like die warpage and localized thermal hotspots threaten both reliability and yield. The shift toward sustainable AI systems for energy-efficient computing requires breakthroughs in everything from hybrid bonding process flows to advanced thermal interface material (TIM) strategies and liquid cooling integration [6].These are not challenges that any single company can solve in isolation. Whether you are a foundry, OSAT, material supplier, or equipment provider, overcoming these bottlenecks requires pre-competitive, industry-wide collaboration. Foundational capabilities must be built collectively before competitive differentiation occurs.This is the core mission of the SEMI Advanced Packaging and Heterogeneous Integration (APHI) Technology Coalition. By collaborating on common standards, shared research frameworks, cross-vendor interoperability models, and collective technology roadmap congruency, APHI is actively dismantling the barriers to next-generation computing.The APHI community is already tackling these issues head-on. Monthly chapter meetings identify and address these and other issues facing heterogeneous integration. The most recent chapter meetings showcased in depth review of these challenges. Jonathan Abdilla from BESI detailed the technical challenges and collaborative research required for global hybrid bonding process flows. Similarly, Dr. Jie Geng from Indium Corporation led a deep dive into crucial TIM strategies for AI and HPC, exploring hybrid stacking evaluation methods and liquid cooling options to combat GPU die warpage.The future of advanced manufacturing will be defined by how effectively we manage power and heat in heterogeneous systems. We invite you to join this critical conversation at the upcoming SEMIEXPO Heartland (April 29-30 in Detroit, MI) Day 2 will feature dedicated sessions on Thermal Management Power Delivery in Advanced Packaging: From TIMs to Warpage Control, as well as strategies for securing the advanced packaging supply chain.To help shape the standards and shared roadmaps that will power the AI revolution, explore our initiatives and get involved with SEMI Advanced Packaging and Heterogeneous Integration (APHI) Technology Coalition.Rafael Tudela is Senior Technical Marketing Manager at SEMI References[1] International Energy Agency (IEA). (2024). Energy and AI Report. [2] U.S. Department of Energy (DOE) Lawrence Berkeley National Laboratory (LBNL). (2024). Report on U.S. Data Center Electricity Demand and Grid Impact.[3] Semiconductor Packaging News. Advanced Packaging and Heterogeneous Integration. Retrieved from: https://www.semiconductorpackagingnews.com/articles/92402.html [4] Yole Group. (2025). Status of the Advanced Packaging Industry 2025.
Read More
As the global semiconductor industry enters a decisive new phase shaped by artificial intelligence, SEMICON Korea 2026 convened the ecosystem from February 11–13 in Seoul, bringing together the companies, technologies, and talent required to sustain momentum on both sides of the AI equation: using AI to transform semiconductor operations, and advancing semiconductor innovation to enable the next generation of AI systems.At the SEMICON Korea press conference, Hyun Cha, President, SEMI Korea stated that with nearly 550 exhibiting companies and over 2,400 booths, the event underscored how progress now depends on a virtuous cycle of collaboration across the entire value chain—from materials and equipment to design, manufacturing, packaging, and systems integration.Opening Perspectives: Collaboration as the Catalyst In the opening ceremony, Ajit Manocha, President and CEO of SEMI, framed the opportunity and challenge ahead: AI is accelerating the industry’s trajectory toward a trillion-dollar market, but sustaining that growth will require deeper collaboration across an increasingly complex ecosystem.That message was reinforced by YH Lee, Chairman of Wonik, who emphasized the need for alignment across the value chain, “looking side to side, not only forward,” as scaling semiconductor technologies grows more difficult. From a policy perspective, Shinhak Moon, Vice Minister of Korea’s Ministry of Trade, Industry Resources (MOTIR), highlighted the importance of the full ecosystem, including parts, materials, and equipment, while cautioning that resilience will be critical amid economic cycles. Together, these perspectives set the stage for a keynote program focused on how AI and semiconductors are now co-evolving.Opening Keynote: Samsung Electronics on Architecting the Future of AI SystemsThe keynote program opened with Jaihyuk Song, Corporate President and CTO of Samsung Electronics, who examined what comes next for AI systems as compute and memory demands rise exponentially. He described a widening gap between compute performance and memory bandwidth, positioning advanced packaging and architectural innovation as central to closing that gap.Song outlined Samsung’s focus on next-generation memory technologies, including high-bandwidth memory and compute-in-memory approaches, as well as the transition beyond traditional Moore’s Law scaling toward planar, vertical, and stacked architectures. His message was clear: sustaining AI performance gains will depend on tight integration across design, process technology, packaging, and system architecture, reinforcing the need for ecosystem-wide coordination. ASE: From Chip Integration to System OptimizationTien Wu, CEO of ASE, expanded the discussion from devices to systems, arguing that advanced packaging has become a primary driver of system-level performance and efficiency. As AI workloads push power, thermal, and bandwidth limits, Wu described a shift from single-chip packages toward heterogeneous integration, 2.5D and 3D architectures, and co-packaged optics.Wu emphasized that productivity, yield, and throughput will increasingly determine competitiveness as packages grow larger and more complex. His perspective reinforced a central theme of SEMICON Korea 2026: AI-driven demand is forcing tighter coupling between design, manufacturing, and packaging, making collaboration not optional, but essential.Cadence: AI-Enabled Design Across the Value ChainBoyd Phelps, Senior Vice President and General Manager of Silicon Solutions at Cadence Design Systems, highlighted how AI is already reshaping semiconductor design and development. As process scaling slows and cost per transistor rises, Phelps described disaggregation and chiplets as a new abstraction layer that enables continued innovation through customization and configurability.He also pointed to the growing role of AI-driven design automation, noting that a significant portion of recent designs leveraged AI-enabled tools. Cadence’s end-to-end portfolio—from IP and tools to packaging and test—illustrated how AI is becoming both a design accelerant and a necessary response to rising system complexity, reinforcing the industry’s virtuous cycle.Lam Research: Velocity Through AI and AutomationThe theme of operational transformation took center stage with Tim Archer, President and CEO of Lam Research, who introduced “velocity” as the defining imperative of the AI era. As AI-driven demand accelerates product cycles and increases complexity, Archer argued that speed must be matched with direction—enabled by AI, automation, and digital twins.Archer detailed Lam’s progress toward autonomous fabs, equipment intelligence, and collaborative virtual development environments that reduce variability and accelerate process development. These capabilities, he explained, allow the industry to respond faster while preserving quality and resilience—another example of AI improving semiconductor operations even as semiconductor innovation enables AI growth.SK hynix: AI as a Tool for Memory InnovationLooking further into the future, Sunghoon Lee, Senior Vice President and Head of R D Process at SK hynix, addressed the mounting difficulty of sustaining memory technology cadence. As stacking, bonding, and material challenges intensify, Lee described a shift toward AI-based R D models that dramatically accelerate material discovery and optimization.By integrating AI into material exploration and process development, SK hynix is shortening development cycles and enabling new memory architectures. Lee emphasized that realizing the full potential of AI-driven R D will require greater data sharing and collaboration across partners—reinforcing the ecosystem-wide virtuous cycle.NVIDIA: From Chips to AI InfrastructureThe final keynote, delivered by Soyoung Jeong, Head of Korea Business at NVIDIA, framed the transformation of NVIDIA from a GPU company into an AI infrastructure provider. He described how accelerated computing and AI factories are reshaping chip design, manufacturing, packaging, and system integration.From AI-assisted design and simulation to system-level optimization and physical AI, NVIDIA’s approach illustrated how semiconductors and AI are now inseparable, each advancing through the other. Partnerships across memory, equipment, and software ecosystems were highlighted as critical to sustaining this momentum.A Program Aligned Around the Same ThemeBeyond the keynotes, SEMICON Korea 2026 reinforced these messages through technology symposia, AI and smart manufacturing forums, cybersecurity discussions, and workforce development initiatives—all focused on enabling AI-powered innovation across the semiconductor lifecycle.Additional Program Highlights: Extending the Virtuous Cycle Across the EcosystemBeyond the keynote stage, SEMICON Korea 2026 reinforced the same virtuous cycle of AI and semiconductor innovation through a wide range of technical, business, and workforce programs designed to engage every layer of the value chain.AI Summit: Translating Strategy into Industrial ImpactThe AI Summit, co‑hosted by SEMI and KAIST, served as a focal point for aligning academic research, device manufacturers, and equipment leaders around AI-powered industrial innovation. Featuring faculty from KAIST alongside representatives from Samsung Electronics, SK hynix, and global equipment companies, the summit explored technology strategies and future roadmaps aimed at accelerating AI adoption across semiconductor manufacturing and design.The discussions reinforced a central theme of SEMICON Korea 2026: AI is no longer an isolated software layer, but a system-level capability that must be embedded across processes, tools, and infrastructure to unlock its full value.Smart Manufacturing Forum: Advancing the Autonomous FabThe Smart Manufacturing Forum highlighted how AI, digital twins, and real-time data are transforming semiconductor fabs toward more autonomous, resilient operations. Speakers shared trends and success cases demonstrating how advanced analytics and AI-driven decision-making are improving yield, productivity, and operational agility.This forum echoed themes raised by equipment and manufacturing leaders in the keynote program, underscoring how AI-driven manufacturing excellence is becoming a prerequisite for meeting the speed, scale, and quality demands of next-generation AI chips.Startup Summit: Fueling Innovation from the Ground UpThe Startup Summit showcased emerging semiconductor and display startups focused on applying AI to improve chip performance, energy efficiency, and manufacturing processes. By connecting startups with industry leaders and venture capital firms—including Applied Ventures, Intel Capital, Samsung Ventures, and SK hynix—the summit emphasized the importance of nurturing innovation across the ecosystem.These early-stage technologies represent the next wave of ideas feeding into the virtuous cycle, where AI-enabled innovation at the startup level can scale rapidly through collaboration with established players.Cybersecurity Forum: Securing the AI-Driven Semiconductor FutureAs AI becomes deeply embedded in semiconductor operations and data flows, the Cybersecurity Forum addressed the growing need for digital trust across the ecosystem. Global experts examined cybersecurity challenges related to compliance, fab security, and AI data governance, highlighting the importance of collaboration to protect sensitive data and intellectual property.The forum reinforced that secure, trusted infrastructure is a foundational requirement for the AI-driven transformation discussed throughout SEMICON Korea 2026.Conclusion: Advancing TogetherSEMICON Korea 2026 made clear that the next phase of industry growth will not be driven by isolated breakthroughs, but by a virtuous cycle of alignment across the full semiconductor value chain. By integrating AI into design, manufacturing, and operations—and by advancing semiconductor technologies that power AI—the industry is building a foundation for sustained innovation. As the event demonstrated, progress will be fastest when the ecosystem moves forward together.Samer Bahou is Director, Corporate Communications at SEMI. Jaegwan Shim is Senior Specialist, Marketing at SEMI Korea.
Read More
The semiconductor industry is hitting a structural inflection point: explosive AI‑driven demand, rapidly rising manufacturing complexity, and stringent sustainability expectations are converging at once. In this context, edge AI deployed directly on tools, sensors, and local controllers, is shifting from experimental to essential, particularly in fabs where milliseconds matter. SEMI’s timely workshop, Smarter Sensors, Smarter Fabs: AI at the Edge in Semiconductor Manufacturing, taking place March 18–19, 2026 in Milpitas, CA, will address this important topic.From sparse sensing to dense instrumentationTwo decades ago, most process tools relied on dozens of sensors per chamber. Today, leading etch, deposition, CMP, and lithography systems routinely integrate hundreds of sensing channels spanning pressure, flow, RF power, optical endpoint, vibration, and chemistry. At 3 nm and 2 nm, process windows are so tight that yield hinges on multivariate understanding of chamber conditions and tool state rather than a few independent alarms. Sensor proliferation has turned fabs into rich data environments—but also exposed the limits of traditional, centrally managed control.Why edge AI is displacing cloud‑only controlConventional architectures push heavy analytics to centralized servers or the cloud, with supervisory systems periodically updating recipes, setpoints, or dispatch rules. Across manufacturing, measured cloud round‑trip times commonly range from 800 to 2,400 ms, whereas edge systems co-located with equipment can respond in 15–45 ms, roughly 50–160× faster. For safety‑ and yield‑critical loops in semiconductor manufacturing, that latency gap is often unacceptable.At the same time, new generations of low‑power neural processing units (NPUs) and edge accelerators deliver tens of trillions of operations per second (TOPS) at single‑digit watt budgets, making always‑on inference viable inside tools, cameras, and controllers. The result is a decisive move toward edge‑native architectures: models execute where data is produced, while cloud resources are reserved for retraining and fleet‑wide learning.Edge AI on the line: control, inspection, and maintenanceIn process control, edge AI is enabling a shift from univariate threshold checks to multivariate models that understand the joint dynamics of sensor streams. Platforms today embed deep‑learning and statistical models directly at or near the tool, performing real‑time endpoint prediction and anomaly detection from high‑dimensional time series. Similar approaches are emerging in lithography and CMP, where local inference helps keep focus, overlay, and removal rate within spec before wafers drift out of control.Inspection and logistics are undergoing a similar transformation. Vision systems with embedded NPUs classify defects at line speed, often above 100 parts per minute, eliminating the need to ship large image volumes to a central cluster. Robots and autonomous mobile robots (AMRs) use local intelligence for short‑horizon planning and collision avoidance, while higher‑level systems focus on global scheduling and optimization.Predictive maintenance is one of the most mature applications: vibration, acoustic, temperature, and pressure data are analyzed locally to detect anomaly signatures hours or days before conventional thresholds trip. Reported benefits include reductions in unplanned downtime, longer component life, and lower maintenance costs when these models are integrated into manufacturing execution systems (MES) and maintenance workflows.Digital twins and agentic AI on top of edge dataDigital twins build on this sensing and edge‑analytics foundation. By maintaining virtual, live‑updated models of tools, lines, and entire fabs, they enable scenario testing, debottlenecking, and root‑cause analysis without putting WIP at risk. Vendors and early adopters report that such twins can shorten process‑node ramps and facility bring‑up by enabling thousands of “what‑if” experiments before physical changes are made.​Agentic AI is now emerging as the orchestration layer above these twins. In semiconductor case studies, agents connected to MES, advanced process control (APC), and planning systems have delivered double‑digit improvements in throughput, cycle time, and tool utilization by autonomously adjusting routing, batch sizes, and scheduling in response to live fab conditions. Other agents mine unstructured engineering notes and fault reports to accelerate root‑cause analysis, turning hard‑won lessons into repeatable, codified behavior.Sustainability as a first‑class requirementSustainability pressures are reinforcing this stack. Semiconductor manufacturing is energy‑ and resource‑intensive, and regulators and customers alike are demanding more transparency and improvement. Edge‑connected monitoring of energy, utilities, and emissions has already helped some fabs cut energy‑related costs by around 20 percent through tighter control of HVAC, process gases, and idle modes. Research initiatives such as imec’s Sustainable Semiconductor Technologies and Systems (SSTS) program are using virtual fab methods and detailed life‑cycle assessment to guide process and equipment choices for lower environmental impact.Strategic takeaways and where to learn moreThe trajectory is clear: fabs that combine dense sensing, edge AI, digital twins, and agentic AI are building toward continuously learning, self‑optimizing operations. Architectures will need to be edge‑first rather than cloud‑only. Simply adding sensors without local intelligence will not deliver competitive advantage, and environmental KPIs are likely to be optimized with the same rigor as yield and cycle time.For practitioners who want to translate these trends into roadmaps, the Smarter Sensors, Smarter Fabs: AI at the Edge in Semiconductor Manufacturing” workshop (March 18–19, 2026, Milpitas, CA) spearheaded by the SEMI Manufacturing Coalitions* will bring together experts in sensing, edge architectures, digital twins, and agentic AI to share concrete deployments and architectures tailored to semiconductor fabs.*The SEMI Manufacturing Coalitions include Smart Manufacturing, Fab Owners Alliance (FOA) MEMS and Sensors Industry Group (MSIG), Advanced Packaging Heterogenous Integration (APHI) and Semiconductor Components, Instruments, and Subsystems (SCIS). Anshu Bahadur is Senior Program Manager, Technology Communities at SEMI. Mark da Silva is Senior Director, Manufacturing Coalitions at SEMI.
Read More
Aki Fujimura has been at the forefront of chip design innovations from the beginning of his career and his technology leadership continues today. He serves as Chairman and CEO of D2S, co-founder of the eBeam Initiative, President of BACUS, and a Governing Council member of the ESD Alliance, a SEMI Technology Community. At Tangent (now Cadence), Fujimura and Steve Teig (a chip designer for the last 20 years and now Vice President and Distinguished Engineer at Amazon) built the first commercial over-the-cell routing system dedicated to fully synchronous designs with timing assurance and automated test-scan insertion. Fujimura and Tom Kronmiller developed LEF/DEF for efficient representation of Manhattan routing, both used as standards in the automated place and route (P R) flow to this day. He again teamed with Teig and Kronmiller to develop the X Architecture, an interconnect architecture based on the pervasive use of 45o diagonal routing. I was thinking about his background as I called him to chat about his evolution from chip design before focusing on chip manufacturing via eBeam technology at D2S.Smith: Let’s talk about your journey from focusing on how to do physical design of chips to chip manufacturing. How did this happen?Fujimura: GPUs weren’t a thing until late 1990s. With CPUs, Manhattan design was the obvious choice for computational efficiency. Largely gridded metal n that went up and down, and metal n+1 that went left and right with vias to connect the line segments were how all automated layout worked. PCB routing and packaging (even back then) used diagonal routing and even curved routing. But chip P R was all Manhattan. That was still true when we worked on the X Architecture at Simplex Solutions (now Cadence). ATi (now inside AMD), NVIDIA and several other GPU companies started in the late 1980s to 1990s, but they were targeting video and gaming more than scientific computing at the time. It’s when Teig came up with the idea for the X Architecture that he wanted to know if 60-degree routing was possible “because a hexagon tessellates a plane.” A good question. I set out to try to find out what the actual limits were in manufacturing that create the limitation to Manhattan shapes. I got introduced to the late Bill Arnold of ASML, who then introduced me to a lot of people in manufacturing who helped me get the answer. Naoya Hayashi of DNP was instrumental in helping me understand that mask making is where the limit exists. Hayashi-san kindly explained to me about the two mask writers. I had to dig around a lot more to make sure that that was the only barrier, but that’s how I came to understand that before masks, everything is data, and after masks, everything is physical. Mask making is the key that enables 45 degrees, but not 60 degrees. The lessons I learned then are still very important to me today. That’s when I saw and appreciated the opportunity there is for software for semiconductor manufacturing.Smith: But you still couldn’t use GPUs for the X Architecture work?Fujimura: Right. Way too early. The idea that GPU-accelerated gaming machines can be connected together to do video editing, or that large scientific simulations can be done on a connected set of gaming machines, was being explored in the 1990s already. It was only 20 years ago (2006) when Jensen Huang announced his bet with the CUDA software stack for general purpose GPUs (GP GPUs) for nodes in racks of CPUs, GPUs, memory and communication to create the modern scientific computer. Six years later in 2012, AlexNet won the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) with CUDA, and the rest is history. But no, we didn’t use GPUs at Simplex. But we did help design GPUs, including with the X Architecture.Editor’s Note: ILSVRC evaluates algorithms for object detection and image classification at large scale. Smith: Now, everything you do at D2S is with GPU acceleration. When and how did that change come about?Fujimura: It was back in 2009, two years after D2S was founded. An extraordinary engineer, Harold Zable, noticed that simulation-based manipulation (rather than rules-based manipulation) of mask shapes, both for wafer manufacturing and for mask manufacturing, would be the ideal application for GPU acceleration. Fast-Fourier Transforms (needed for lithography simulation and optical proximity correction (OPC)/inverse lithography technology (ILT)) and Gaussian manipulations (needed for eBeam mask simulation and mask process correction (MPC) are nearly “free” in terms of compute time on GPUs. You still have to get the data in and out efficiently, but you can do pretty sophisticated computing without much overhead. At the same time, multi-beam based eBeam writing was getting momentum, first in wafer direct write applications. In 2007, at the BACUS conference in Monterey, Calif., IMS—then a well-respected research organization in Vienna—published a paper saying that multi-beam for mask writing is what they’d like to do. The wafer market is much bigger, but this technology is more suited for mask writing, where write times are measured in hours per mask. “Wafers Per Hour” is the measure in wafer manufacturing, so mask writing gets to flip the division. We were looking at a mask design and mask manufacturing world that should be doing simulation-based manipulation rather than rule-based. That’s better with GPUs. On top of that, maybe the world is going to go to multi-beam writing, going away from four decades of variable-shaped beam (VSB) writing. And I knew from the X Architecture experience that VSB was the only thing in the eco-structure that restricted mask shapes to be Manhattan or 45 degrees. In fact, with multi-beam, any curvilinear shape within the limits of resolution of a given pixel size can be freely written on the mask. The only barrier then to having curvilinear masks would be the software stack and trying to compute it with CPUs only. We knew GPU acceleration was the answer. Smith: Was it just totally an accident that multi-beam and GP GPUs happened at the same time?Fujimura: Yeah, it was. However, just as when multiple people simultaneously invent the same thing without knowing about each other, the environment and times in which we live have a lot to do with this. So, I guess, it’s not really just “luck.” But GP GPUs in 2006 and IMS Multibeam in 2007, I think that’s luck.Anyway, D2S became the GPU-acceleration partner for the semiconductor manufacturing industry and decided to work only on things that can be accelerated by GPUs in 2012.Smith: What trends do you see going forward in the next three to five years?Fujimura: A move toward curvilinear mask features, as well as an increased interest in curvilinear wafer targets as designers become aware that the manufacturing side has established a solid path for curvilinear mask shapes. We’re leaving a lot of margin on the table to accommodate gridded Manhattan assumptions, and that’s really no longer necessary from a manufacturing standpoint. I think electronic design automation (EDA) should be working on enabling curvilinear designs, because the door is open for the design world to explore curvilinear chip design and to reap compelling benefits in terms of power/performance and reliability.Editor’s Note: While Manhattan geometries are rectilinear shapes aligned to vertical and horizontal axes, curvilinear design introduces smooth, continuous curves into layouts and masks, leveraging advanced computational lithography and mask-writing technologies. This improves pattern fidelity, electrical performance and manufacturability at advanced technology nodes.About Aki FujimuraAki Fujimura is chairman and CEO of D2S, Inc., and managing company sponsor of the eBeam Initiative. Previously, Fujimura was CTO at Cadence Design Systems, President/COO and inside board member of Simplex Solutions, and VP and inside board member at Pure Software. He co-founded Tangent Systems (acquired by Cadence).Fujimura, made a SPIE fellow in 2023, serves as President of the SPIE BACUS Technical Group. He serves on the governing council of the ESD Alliance, a SEMI Technology Community. Fujimura was on the board of HLDS, RTime, Bristol, S7, and Coverity, Inc.Fujimura received his BSEE and MSEE degrees from MIT.Robert (Bob) Smith is executive director of the ESD Alliance, a SEMI Technology Community. 
Read More