In recent years, Intel has embarked on a bold and transformative journey to reclaim its position as a leader in semiconductor technology. This ambition is embodied in the creation and expansion of its Foundry business division, a strategic pivot aimed at not only recapturing lost ground against rivals like Taiwan Semiconductor Manufacturing Company (TSMC) and Samsung Electronics but also establishing Intel as a pivotal player in the global chip manufacturing landscape.

Despite facing significant financial challenges, Intel’s leadership under CEO Patrick Gelsinger has laid out a comprehensive roadmap for resurgence and growth.

Intel’s strategic pivot in the semiconductor space

The operational challenges facing Intel’s Foundry business have been stark. In 2023, the division reported operating losses of US$7 billion, a notable increase from US$5.2 billion the year prior, alongside a 31% drop in revenue. These figures, reflective of the intense competition and rapid technological evolution in the semiconductor industry, have prompted a strategic reassessment at Intel, as noted by Reuters.

CHINA’S NEW TECH POLICIES CHALLENGE INTEL AND AMD IN A SHIFTING LANDSCAPE

Muhammad Zulhusni | 26 March, 2024

Amid these financial headwinds, Intel has undertaken significant structural and strategic shifts. One of the most critical moves has been the separation of manufacturing costs from the broader product development expenses on the income statement, creating a distinct category for Intel Foundry. This accounting realignment aims to provide a clearer picture of the foundry’s performance and the impact of production costs on Intel’s overall financial health.

At the heart of Intel’s strategy is a deep commitment to technological innovation and excellence. The company has announced ambitious plans to introduce five semiconductor process nodes over four years, a rapid pace of development aimed at leapfrogging competitors. A key focus is the advanced 18A technology, positioned as Intel’s counter to TSMC’s 3-nanometer node. With strong demand and several customer commitments already in place, the 18A technology represents a significant bet on Intel’s future in chip manufacturing.

Intel’s pivot to extreme ultraviolet (EUV) chip manufacturing technologies marks another critical aspect of its strategy. After initial hesitations that contributed to operational losses and competitive disadvantages, the adoption of EUV technology is expected to dramatically improve cost efficiency and product performance. This shift addresses past missteps and positions Intel favorably in the “post-EUV era,” as CEO Gelsinger highlighted.

Despite the current challenges, Intel’s outlook is optimistic. The company has outlined a path to break even by 2027, driven by efficiencies from EUV technology, reduced reliance on external contract manufacturers, and a growing portfolio of customer commitments. Intel’s ambitious investment of US$100 billion in expanding U.S. chip manufacturing facilities underscores its commitment to recovery and long-term dominance in the foundry space.

Intel’s US$7 billion loss spurs semiconductor industry reinvention (Source – X)

Expanding manufacturing footprint with federal support

Regarding this investment strategy, Reuters reported that Intel is embarking on an expansive effort to construct and enhance factories across four states in the U.S., buoyed by US$19.5 billion in federal grants and loans. Furthermore, the company is pursuing US$25 billion in tax incentives.

The focal point of this five-year initiative is to transform sprawling fields near Columbus, Ohio, into the world’s most significant AI chip production facility by as early as 2027, as explained by CEO Pat Gelsinger.

This influx of federal support comes courtesy of the CHIPS Act, underscoring the U.S. government’s commitment to revitalizing domestic semiconductor manufacturing.

Intel’s expansive blueprint includes upgrades to New Mexico and Oregon facilities and broadening its footprint in Arizona. This move comes as TSMC is also escalating its presence in Arizona, encouraged by President Joe Biden’s efforts to reinvigorate advanced semiconductor production in the U.S.

The financial aid from the Biden administration’s strategy to foster a semiconductor manufacturing renaissance is crucial for Intel’s recovery. Once a leader in producing the most advanced and compact semiconductors, the company saw its dominance wane in the 2010s as it fell behind TSMC, resulting in reduced profit margins from lowered prices to maintain market share with fewer products.

In 2021, Gelsinger unveiled a strategy to reinstate Intel’s position at the forefront of semiconductor manufacturing, acknowledging the need for governmental support to ensure profitability.

With this governmental backing secured, Intel is set to channel about 30% of its US$100 billion investment into construction, covering labor, piping, and concrete expenses. The remainder will finance the acquisition of chip-making equipment from companies like ASML, Tokyo Electron, Applied Materials, and KLA.

This equipment is pivotal for activating the Ohio site between 2027 and 2028, although Gelsinger cautioned that this timeline might extend if the semiconductor market faces downturns. Despite relying on grants and loans, Intel aims to fund most equipment purchases through its existing cash reserves.

Gelsinger has voiced that additional U.S. financial support for chip factories is essential for reasserting the country’s semiconductor manufacturing leadership, a sentiment he reiterated recently.

Despite federal support, Intel faces pressure to promptly demonstrate its competitiveness against Taiwanese and Korean rivals, according to Ben Bajarin, CEO of Creative Strategies.

INTEL UNLEASHES AI CHIPS FOR PCS AND DATA CENTERS, CHALLENGING AMD AND NVIDIA

Dashveenjit Kaur | 18 December, 2023

Vision 2030: Intel’s Foundry ambition

The strategic realignment towards its Foundry business and aggressive technological advancements reflect Intel’s broader vision. By 2030, Intel aims to become the world’s second-largest foundry, boasting operating margins driven by innovation, cost-efficiency, and a robust client base. This vision is buoyed by the current lifetime value of contracts for the Foundry business, which stands at US$15 billion, and the anticipated benefits of transitioning to EUV manufacturing technologies.

Intel’s journey through financial turbulence and strategic shifts in its Foundry business division is a testament to its resilience and determination to regain its leadership in the semiconductor industry. Through technological innovation, strategic restructuring, and financial investment, Intel is not just aiming to navigate its current challenges but positioning itself for a future where it leads the next wave of semiconductor manufacturing.

As the industry continues to evolve rapidly, Intel’s efforts to redefine its operational and strategic paradigms will be closely watched by competitors and partners alike, marking a new chapter in the global semiconductor landscape.

The post Intel reinvents in semiconductor industry amid US$7 billion loss appeared first on Tech Wire Asia.

Elon Musk has declared that his artificial intelligence startup xAI will extend access to its chatbot Grok to all premium subscribers of the social media platform X. This announcement, made in a post on X, does not delve into further details but signifies a shift from the chatbot’s previous limitation to Premium+ subscribers. Amidst advertisers withdrawing from X, Musk is pivoting away from advertising revenue, focusing instead on enhancing subscription services.

In a move that critiques the profit-oriented use of technology by major tech firms such as Google, Musk plans to make Grok open-source. This follows his lawsuit against OpenAI, where he accuses the organization of straying from its non-profit roots towards profit-driven motives. By open-sourcing Grok, Musk aligns xAI with entities like Meta and France’s Mistral, which have also made their AI models publicly available, encouraging innovation and engagement from the wider community.

Elon Musk posted the availability of Grok for all premium subscribers (Source – X)

What’s going on with Elon Musk and the world of AI development?

Moreover, Musk’s legal confrontation with OpenAI, which he co-founded and eventually left, highlights his concerns about the ethical trajectory of AI development. Despite previously endorsing profit-focused strategies, including a merger proposal with Tesla, Musk’s latest initiatives and comments, especially at the AI Safety Summit in the UK, advocate for ethical AI development and the adoption of open-source principles, with the aim of developing a “maximum truth-seeking AI” at xAI.

This strategic direction not only challenges the methodologies of OpenAI and Google but also ignites a debate among technology leaders and investors about the implications of making AI technology open-source. While such transparency can foster innovation, there are concerns about its potential misuse, underscoring the balance between technological advancement and ethical considerations.

As reported by BBC, Elon Musk’s Neuralink has showcased its first patient, who, using a brain implant, controlled a computer cursor and played online chess. In a nine-minute live stream on X, viewers witnessed Noland Arbaugh, paralyzed below the shoulders due to a diving accident, using the device. Arbaugh, who received the chip in January, described the surgery as “super easy.”

A demonstration of the controlled a computer cursor and played online chess through the brain (Source – X)

Arbaugh also recounted playing the video game Civilization VI for eight hours straight, facilitated by the brain implant, though he mentioned encountering some issues with the technology. The Neuralink device, about the size of a one-pound coin, is designed to be inserted into the skull, with tiny wires that can read neuron activity and send wireless signals to a receiver.

Following trials in pigs and demonstrations of monkeys playing a basic version of Pong, the FDA approved Neuralink for human testing in May 2023. Neuralink is among a growing number of firms and academic departments pushing the boundaries of brain-computer interface (BCI) technology.

In a parallel development, the École Polytechnique Fédérale in Lausanne, Switzerland, enabled paralyzed individual Gert-Jan Oskam to walk by simply thinking about moving, using electronic implants on his brain and spine that wirelessly relay thoughts to his legs and feet, as reported in Nature.

BCIs aim to capture some of the electrical impulses generated by the brain’s approximately 86 billion neurons, which facilitate movement, sensation, and thought. These impulses can be detected by non-invasive caps or directly via implanted wires, drawing significant research investment.

Musk has boldly claimed that Neuralink’s technology can restore sight in monkeys, branding this technology as “Blindsight.” He envisions this technology, initially offering low-resolution vision akin to early video games, eventually surpassing human visual capabilities. He assures that the procedures have been safe for the animals involved.

TWITTER DATA BREACHES EXPOSE WEAKNESSES ELON MUSK NEEDS TO SOLVE QUICKLY

Aaron Raj | 9 January, 2023

Neuralink’s advancements, including the “Telepathy” product enabling mind-controlled computer use, mark significant strides in the field. Following FDA approval for its first human trial, Neuralink released a video showing a quadriplegic patient playing chess through mind control, demonstrating the implant’s potential through 64 flexible threads that record and transmit brain signals.

Challenges and advances: Neuralink’s journey to human trials

According to Reuters, a U.S. health policy lawmaker has queried the FDA about its prior inspection of Neuralink, before approving it for human trials. This follows reports of issues discovered during inspections related to animal testing practices at Neuralink. These findings emerged shortly after Neuralink announced FDA clearance for human testing of its brain implants, which enable paralyzed individuals to control computers with their minds.

Representative Earl Blumenauer expressed concerns in a letter to the FDA about overlooked evidence from animal testing violations dating back to 2019. He questioned how the FDA reconciled these reports with its decision to authorize Neuralink’s human trials, amidst allegations of rushed experiments leading to unnecessary animal suffering and potential data integrity risks.

In response, the FDA indicated it would directly address Blumenauer’s inquiries, noting its post-approval inspection did not identify any significant safety concerns for the trial. Neuralink, having commenced human testing, recently demonstrated the implant’s capabilities in a live stream, highlighting the potential of brain-computer interfaces despite regulatory and ethical scrutiny.

This exploration into brain-computer interface technology, with companies like Synchron and Blackrock Neurotech also advancing in human trials, demonstrates the potential for patients to control digital interfaces solely through thought. The investigation into Neuralink’s regulatory approval underscores the critical balance between innovation in medical technology and the necessity of maintaining safety and ethical standards.

As the field progresses, the discussions around Neuralink’s practices and the FDA’s oversight reflect broader questions about the pace of technological advancement and the frameworks needed to ensure its responsible development.

The scrutiny faced by Neuralink, arising from concerns over its animal testing procedures and the subsequent approval for human trials, brings to light the challenges of pioneering medical devices within the rapidly evolving domain of brain-computer interfaces. It emphasizes the importance of rigorous regulatory processes that not only facilitate technological breakthroughs but also safeguard the welfare of both animal subjects and human participants.

The post Elon Musk takes AI and brain chips mainstream with Grok and Neuralink appeared first on Tech Wire Asia.

At the Consumer Electronics Show (CES) 2024, announcements in the automotive industry are always most highly anticipated. Apart from tech companies unveiling their latest innovations in various products and services, carmakers also showcase new models and features. But the automotive industry is just one part of CES 2024.

As expected, AI was center stage at this year’s CES, with most of the new products built on the technology’s use cases. That includes self-driving baby carriages, chatbots in cars as well as AI in beauty products. Among the big names participating were European and Chinese carmakers, mobile phone companies, and chip companies.

CHINESE EVS PICK NVIDIA DRIVE FOR AUTOMATED DRIVING

Aaron Raj | 10 January, 2024

Nvidia, which dominated most of the announcements at CES 2024, unveiled several new products for gaming laptops as well as new partnerships with EV carmakers in China. But apart from Nvidia, two other chip companies are hoping to not only make an impact at CES 2024 but also challenge Nvidia across the year with their products.

Intel and Qualcomm Technologies have long dominated the mobile and laptop market. The chips produced by these two companies continue to enable innovation in these devices as well. And with AI adoption increasing, both Intel and Qualcomm are hoping to play an important role in developing the technology with a big focus on the automotive industry.

Qualcomm Technologies announced breakthrough innovations across its Snapdragon Digital Chassis portfolio.

Qualcomm continues to innovate in the automotive industry

At CES 2024, Qualcomm Technologies announced breakthrough innovations across its Snapdragon Digital Chassis portfolio.  The company is in its twentieth year of supplying technology for the automotive industry, with revenue growing by double digits year-over-year, which has been driven by the increased adoption of Snapdragon Digital Chassis solutions.

“Our commitment to advancing automotive technology in support of global automakers, Tier-1 suppliers and our ecosystem partners is helping shape the future of software-defined vehicles and is accelerating us into a new era for the automotive industry,” said Nakul Duggal, SVP and general manager, automotive & cloud computing, Qualcomm Technologies, Inc.

Some of the offerings by Qualcomm for the automotive industry. (Image – Qualcomm).

Some of the innovations announced include:

• Snapdragon Auto Connectivity Platform – the roadmap developed through Qualcomm’s decades of connectivity leadership. Automakers are being equipped to meet the increasing needs for higher levels of safety and intelligence powered by LTE, 5G, connected services, vehicle-to-everything (V2X), wi-fi, bluetooth, satellite communications and precise positioning.
• Snapdragon Cockpit Platform – the platform helps automakers create immersive, intuitive, and sophisticated in-vehicle experiences with its enhanced graphics, multimedia and AI capabilities.
• Snapdragon Chassis Platform – Qualcomm welcomes the era of AI for automotive and is using its industry-leading AI hardware and software solutions for the Snapdragon Digital Chassis Platform to help drive automotive AI forward. The Snapdragon Digital Cockpit Platforms are currently available with generative AI capabilities.
• Snapdragon Ride Platform – one of the automotive industry’s most advanced, scalable and customizable automated driving system-on-chip (SoC) family that is designed to help global automakers and Tier-1 suppliers build efficient automated driving (AD) solutions.
• Snapdragon Ride Flex – high-performance central compute SoCs that are designed to support mixed-criticality workloads across heterogenous compute resources, allowing for digital cockpit, advanced driver-assistance system (ADAS) and AD functions to co-exist on a single SoC.
• Snapdragon Car to Cloud – allows new features and services to be added throughout the vehicle’s lifecycle for highly personalized experiences across all tiers – letting automakers and fleet providers stay directly connected to consumers beyond the point of sale.

Qualcomm is also offering the Snapdragon Digital Chassis SoCs for two-wheelers and new vehicle class segments. The Digital Chassis SoCs are designed to enhance safety and experiences for the end user, offering fully integrated solutions that bring connectivity, infotainment, advanced rider assistance systems (ARAS) and personalized, cloud-connected digital services to motorcycles, ICE and electric scooters, 3-wheelers, e-bikes, ATVs, and vehicles for farming and agriculture.

Qualcomm is also offering the Snapdragon Digital Chassis SoCs for two-wheelers and new vehicle class segments. (Image – Qualcomm).

Watch out Nvidia, Intel has big plans to take on the industry

CARS OF THE FUTURE: HOW VIRTUAL TWIN TECHNOLOGY IS REINVENTING AUTOMOTIVE DESIGN

Aaron Raj | 2 November, 2023

Intel also has big plans for the automotive market. At CES 2024, Intel announced that it plans to acquire Silicon Mobility, a fabless silicon and software company that specializes in SoCs for intelligent EV energy management. Silicon Mobility’s SoCs feature industry-leading accelerators purpose-built for energy delivery and co-designed with highly advanced software algorithms for significant gains in vehicle energy efficiency.

Last year, Intel sold off a small stake in Mobileye, a self-driving technology company. Should the Silicon Mobility acquisition go through, it will extend Intel’s reach in the vehicle beyond high-performance compute into intelligent and programmable power devices.

“Intel is taking a ‘whole vehicle’ approach to solving the industry’s biggest challenges. Driving innovative AI solutions across the vehicle platform will help the industry navigate the transformation to EVs. The acquisition of Silicon Mobility aligns with our sustainability goals while addressing a critical energy management need for the industry,” said Jack Weast, vice president and general manager of Intel Automotive.

Andy An (left), president of Geely Holding Group and CEO of Zeekr Intelligent Technology, greets Intel CEO Pat Gelsinger (center) and Jack Weast, vice president and general manager of Intel Automotive, on Tuesday, Jan. 9, 2024, at CES in Las Vegas. (Credit: Intel Corporation).

Given that Nvidia has grown its influence in the EV industry, apart from the acquisition, Intel also has a new family of AI-enhanced software-defined vehicle system-on-chips (SoCs). China’s Zeekr will be the first original equipment manufacturer (OEM) to adopt the new SoC to deliver its generative AI-driven living room experiences to next-generation vehicles.

The new family of AI-enhanced SDV SoCs addresses a critical industry need for power and performance scalability. The family of SoCs features AI acceleration capabilities from Intel’s AI PC roadmap to enable the most desirable in-vehicle AI use cases, such as driver and passenger monitoring.

“Intel’s AI-enhanced SDV SoCs combine the best of AI PC and Intel data center technologies necessary to support a true software-defined vehicle architecture,” added Weast.

To ensure the strict quality and reliability requirements necessary for automotive use cases, Intel plans to work with R&D hub Imec. The R&D company will ensure Intel’s advanced chiplet packaging technologies meet the requirements set. The move underscores a commitment to be the first automotive supplier to support the integration of third-party chiplets into its automotive products.

Intel SoCs are in more than 50 million vehicles, powering infotainment, displays, digital instrument clusters and more. Intel’s expanded AI-enhanced “whole vehicle” roadmap will move the industry toward a more scalable, software-defined and sustainable future.

The post CES 2024: Intel and Qualcomm to take on Nvidia in automotive industry appeared first on Tech Wire Asia.

Nvidia is spearheading the integration of AI into its chip design processes by incorporating chatbots, a strategy reflecting the larger tech industry’s shift towards more AI-centric solutions. This initiative shows Nvidia’s commitment to AI (and to staying ahead of the pack), and represents a pivotal change in how AI can simplify and streamline complex engineering tasks.

By introducing chatbots into its chip design framework, Nvidia aims to refine its design methodologies and establish a new precedent in the application of AI in technology development. The company is betting that using AI to design smarter chips will significantly influence the future of AI-assisted engineering.

Nvidia’s recent research is focused on using chatbots capable of generating conversational, human-like responses in semiconductor design. This research illuminates how companies in niche sectors can tailor large language models (LLMs) to their own datasets, creating specialized assistants that enhance operational efficiency.

THE NVIDIA EUREKA MASTERS HUMAN-LIKE PEN-SPINNING TRICKS, REDEFINING AI LEARNING

Muhammad Zulhusni | 24 October, 2023

The AI revolution in semiconductor design

Semiconductor design is an extraordinarily complex endeavor. The intricate architecture of state-of-the-art chips like Nvidia’s H100 Tensor Core GPU can be compared to a densely populated city under a microscope, consisting of billions of transistors, each a fraction of the width of a human hair.

The design of modern microchips, involving the strategic placement of tens of billions of transistors on a silicon wafer, is among the most challenging tasks in the tech industry. To construct such a digital metropolis, diverse engineering teams often collaborate for up to two years. These teams are responsible for various tasks ranging from the overall architectural blueprint, to crafting microscopic circuits, and thoroughly testing their functionalities. Each of these roles demands specialized methods, software, and computer languages.

Nvidia’s chips, recognized for their complexity, are pivotal in powering advanced technologies, including AI systems like ChatGPT. This complexity necessitates continual innovation and precision in design and execution.

The study conducted by Nvidia discovered that incorporating extensive, specific internal data enabled even basic chatbots to outperform their advanced counterparts in accuracy. This approach not only optimizes performance but also helps manage and control system costs.

How AI can benefit engineering

Nvidia’s using AI to design new chips.

One of the standout features demonstrated by Nvidia is the use of AI in generating code. Bill Dally, Nvidia’s chief scientist, pointed out that engineers often spend considerable time identifying and diagnosing faults in chip components. AI systems can alleviate this burden by rapidly writing code scripts and facilitating testing.

Mark Ren, an Nvidia research director, believes that in the future, large language models will be instrumental across all facets of semiconductor design.

“This is an important first step in applying LLMs to the complex work of designing semiconductors,” said Dally. He emphasized the feasibility of highly specialized sectors using their own data to train effective AI models.

FOXCONN AND NVIDIA COLLABORATE TO SUPERCHARGE THE AI INDUSTRY

Dashveenjit Kaur | 19 October, 2023

Nvidia’s engineers have created ChipNeMo, a bespoke LLM tailored for their specific design needs, trained on the company’s confidential data. ChipNeMo is a pioneering project investigating the potential applications of LLMs in industrial chip design.

The team opted for unique domain adaptation techniques rather than using generic LLMs. This included deploying custom tokenizers and domain-specific training to enhance the model’s performance in specific applications like engineering assistance, script generation, and bug analysis.

These techniques have already shown promising results, significantly improving the model’s efficacy and suggesting possible reductions in model size while maintaining or enhancing performance in various design tasks.

Nevertheless, a noticeable disparity remains between the current achievements and the ideal outcomes. Nvidia’s team is confident that further exploring domain-adapted LLMs could help bridge this gap, leading to even more sophisticated and efficient chip design methodologies.

Looking ahead, Nvidia engineers want to apply generative AI throughout the chip design process, potentially unlocking substantial improvements in productivity and innovation. With a career spanning over two decades in electronic design automation (EDA), Ren foresees this AI integration as a game-changer in the semiconductor industry.

Among the innovations, the bug maintenance tool has drawn significant acclaim. This tool, which automates the upkeep of known bug descriptions, has proved invaluable in streamlining the debugging process.

A prototype chatbot designed to respond to questions regarding GPU architecture and design showed potential in early tests, letting engineers access pertinent technical documents rapidly. In development, a code generator already produces short snippets of software in two languages used by chip designers. This generator is set to be integrated into current design tools, offering engineers a convenient assistant for ongoing projects.

ChipNeMo in action. (Source – Nvidia)

Nvidia and the future of AI in chip development

The research paper essentially discusses the team’s endeavor to gather and utilize Nvidia’s design data to construct a specialized generative AI model. While initially focused on chip design, this process has implications far beyond, indicating its potential applicability across various industries.

The project began with a base model, further developed using Nvidia NeMo. The foundational model boasts 43 billion parameters, indicative of its pattern recognition capabilities, and was trained using over a trillion tokens, comprising words and symbols from texts and software.

ChipNeMo exemplifies how a specialized technical team can enhance a pre-existing model using their unique data. (Source – Nvidia)

Subsequent training phases involved around 24 billion tokens of Nvidia’s design data, followed by a mixture of approximately 130,000 examples of design conversations and layouts.

IS NVIDIA ABUSING ITS DOMINANCE IN THE AI MARKET?

Dashveenjit Kaur | 2 October, 2023

This pioneering work is among the early instances of research and proof-of-concept demonstrations of generative AI in the semiconductor industry, signaling the beginning of a new era in technological innovation and AI application. As this field continues to evolve, it’s clear that Nvidia’s contributions and explorations will not only reshape the landscape of chip design but also illustrate the vast, untapped potential of AI across diverse sectors.

The post How Nvidia is using AI chatbots to craft smarter chips appeared first on Tech Wire Asia.

While there are many companies in the semiconductor manufacturing market, some components are the speciality of only a handful of companies. In fact, the semiconductor manufacturing process can be divided into two main stages – design and manufacturing.

Companies that focus only on design are referred to as fabless firms. These include companies like Qualcomm, Nvidia and AMD. Companies that focus on manufacturing are called foundries. Examples of foundries include GlobalFoundries, TSMC, Micron Technologies and SMIC. There are also semiconductor companies that do both. They are called integrated device manufacturers (IDMs). These companies include Intel, Samsung and Texas Instruments.

NVIDIA ACHIEVES A BREAKTHROUGH IN CHIP MANUFACTURING FOR ASML, TSMC, AND SYNOPSYS

Dashveenjit Kaur | 23 March, 2023

And then there is the semiconductor manufacturing process supply chain, which involves components from all over the world to make a chip. The semiconductor supply chain remains the most crucial element in the entire process.

The Covid-19 pandemic witnessed unprecedented disruptions to the semiconductor supply chain. Since then, companies have been working to solve their supply chain issues by finding alternatives to speed up the manufacturing process. Some companies have even started to develop their own chips to deal with the issue.

But apart from the pandemic, the trade conflict between the US and China has also affected the production of chips. China is one of the largest manufacturers of semiconductor components for American companies. At the same time, a lot of American brands also have plants operating in China to build their products.

Canon’s new innovation.

Sanctions by the US government on the export of semiconductor materials to China continue to be a concern for most companies. One company caught in this crossfire is ASML. The Dutch company is known for its immersion deep ultraviolet lithography machines, which are a key component of the semiconductor manufacturing process.

Under the restrictions by the US, Dutch export control rules will forbid also ASML from maintaining, repairing and providing spare parts for controlled equipment without specific government approval.

According to a report by The Diplomat, ASML is the world’s largest producer of photolithography machines, which use a light source to etch electronic pathways onto silicon wafers. This process is an essential step in creating the semiconductors that are sold by the likes of TSMC, Intel and Nvidia, which all purchase ASML machines for their chipmaking.

The report also stated that ASML is currently the only company in the world that can manufacture extreme ultraviolet (EUV) machines. Its EUVs are sold for up to US$200 million per machine. The newest model in production, the High NA EUV machine, is valued at over US$300 million and is roughly the size of a truck.

An FPA-1200NZ2C in operation. (image by Canon)

Canon looks to boost semiconductor manufacturing process

Now, Canon seems to have made a breakthrough with lithography machines. On October 13^th, Canon launched the FPA-1200NZ2C nanoimprint semiconductor manufacturing equipment. The machine is capable of executing circuit pattern transfer, the most important semiconductor manufacturing process.

Canon has already been involved in the semiconductor manufacturing process with its existing photolithography systems. By introducing nanoimprint lithography (NIL) technology, Canon will be able to reproduce intricate circuit patterns on the wafer without going through an optical mechanism. It allows for the creation of complex two- or three-dimensional circuit patterns in a single imprint, potentially reducing the cost of ownership.

WHERE DOES ASML ACTUALLY STAND IN THE US-CHINA CHIP FEUD?

Dashveenjit Kaur | 27 January, 2023

Canon’s NIL technology achieves a minimum linewidth of 14 nm, meeting the requirements for producing advanced 5-nm-node logic semiconductors. With advancements in mask technology, it is anticipated that this technology could enable circuit patterning with a minimum linewidth of 10 nm, equivalent to a 2 nm node. The new product incorporates state-of-the-art environment control technology, minimizing contamination by fine particles within the equipment.

Put simply, this technology enables the precise alignment crucial for manufacturing semiconductors with several layers, reducing defects caused by fine particles. It facilitates the creation of intricate circuits, contributing to the production of cutting-edge semiconductor devices. Unlike current photolithography equipment, this new product doesn’t require a special wavelength light source for fine circuits, significantly reducing power consumption and contributing to CO2 reduction.

Additionally, Canon says that the FPA-1200NZ2C can be applied to various uses, including metalenses for XRs, featuring microstructures at the nanometer scale, in addition to logic and other semiconductor devices.

Is there hope for China’e semiconductor manufacturing process? (Image by Shutterstock)

Hope for China’s semiconductor industry?

Since the sanctions imposed by the US, China has already been working round the clock on developing its own equipment. The world’s most populous country has set itself a target of reaching 40% self-sufficiency in chips by 2020 and 70% by 2025. However, it has not even passed the 20% mark yet.

Despite research and developments in local Chinese universities, the country is still struggling to be self-sufficient and continues to import around US$400 billion worth of semiconductors annually.

But according to a report by Bloomberg, Canon’s new machine may open up a new front in the US-China trade war. Currently, the import of EUV machines into China is prohibited by trade sanctions. And until now, importing EUV machines has been the only reliable way to domestically fabricate 5nm chips and smaller.

The Japanese firm’s technique skips photolithography altogether and instead impresses the desired circuit pattern onto the silicon wafer. Because of its novelty, Bloomberg believes that it’s unlikely to be expressly forbidden by existing trade curbs.

That could drive a tree trunk through the US’ sanction plans – or at least force it to steer another set of openly Sinophobic restrictions into place.

Things could be about to get interesting in the world of semiconductor manufacturing again.

The post Canon gives China hope in semiconductor manufacturing process appeared first on Tech Wire Asia.

The rivalry between Advanced Micro Devices, Inc (AMD) and Nvidia Corp dates back decades — even before the former bought ATI for its graphics division. Today, they’re still battling it out as the top two graphic cardmakers for gaming PCs, but that’s not where the rivalry ends.

Both companies’ primary focus is now shifting to AI, and the competition has never been more fierce. Although Nvidia dominates the market today with its AI chips, AMD is becoming a strong challenger as it does more to intensify the rivalry.

Nvidia holds a dominant position in the market for high-performance chips used in the development of ChatGPT and the AI services that have surged in popularity across the tech industry in 2023. That surge in demand following ChatGPT’s frenzy has propelled Nvidia’s market capitalization beyond US$1 trillion. But the surge was so strong and so sudden that the need for Nvidia’s AI chips has led to a supply shortage, which the company has acknowledged and is actively addressing.

SHOULD WE STILL CARE ABOUT THE CPU COMPARISON BETWEEN AMD, INTEL, AND NVIDIA?

Muhammad Zulhusni | 19 September, 2023

AMD is using the shortage to actively narrow the AI gap on Nvidia, announcing its most recent move this week. AMD is acquiring open-source AI software company Nod.ai – a move made to expand the company’s open AI software capabilities. The US-based chip giant said the agreement strongly aligns with its AI growth strategy centered on an open software ecosystem that lowers the barriers of entry for customers through developer tools, libraries, and models.

“The addition of Nod.ai will bring an experienced team that has developed an industry-leading software technology that accelerates the deployment of AI solutions optimized for AMD Instinct™ data center accelerators, Ryzen™ AI processors, EPYC™ processors, Versal™ SoCs and Radeon™ GPUs to AMD,” the company explained.

AMD improves as a prospect with the acquisition of Nod.ai. Source: X

Nod.AI, or Nod Labs, builds open-source technologies “for future AI systems,” according to the startup. It mainly specializes in reinforcement learning, a type of system that “learns” via trial and error. The acquisition is expected to close this quarter, an AMD spokesperson told CNBC.

“At Nod.ai, we are a team of engineers focused on problem-solving — quickly – and moving at pace in an industry of constant change to develop solutions for the next set of problems,” said Nod.ai’s co-founder and CEO Anush Elangovan. “By joining forces with AMD, we will bring this expertise to a broader range of customers on a global scale.”

DANA POINT, CALIFORNIA – SEPTEMBER 26: Dr. Lisa Su, Chair and CMO, AMD speaks onstage during Vox Media’s 2023 Code Conference at The Ritz-Carlton, Laguna Niguel on September 26, 2023 in Dana Point, California. Jerod Harris/Getty Images for Vox Media/AFP (Photo by Jerod Harris / GETTY IMAGES NORTH AMERICA / Getty Images via AFP)

Adding to Elangovan’s context, Nod.ai delivers optimized AI solutions to top hyperscalers, enterprises, and startups. The compiler-based automation software capabilities of Nod.ai’s SHARK software reduce the need for manual optimization and the time required to deploy highly performing AI models to run across a broad portfolio of the data center, edge and client platforms powered by AMD CDNA™, XDNA™, RDNA™ and “Zen” architectures.

AMD believes that acquiring Nod.ai will significantly enhance its ability to provide AI customers with open software to deploy highly performing AI models tuned for AMD hardware quickly. Vamsi Boppana, senior vice president of the Artificial Intelligence Group at AMD, said, “The addition of the talented Nod.ai team accelerates our ability to advance open-source compiler technology and enable portable, high-performance AI solutions across the AMD product portfolio.”

Boppana sees an advantage in the sheer fact that Nod.ai’s technologies are already widely deployed in the cloud, at the edge, and across a broad range of endpoint devices today. The news of the acquisition came on the heels of the company’s purchase of French startup Mipsology to strengthen its AI inference software capabilities.

On August 24, the California-based chip giant unveiled the acquisition, saying that Mipsology, “a leader in AI software and long-standing AMD partner,” will help the company “accelerate our customer engagements and expand our AI software development capabilities.” Boppana, in a statement, wrote, “The team will help develop our full AI software stack, expanding our open ecosystem of software tools, libraries, and models to pave the way for streamlined deployment of AI models running on AMD hardware.”

AMD VS NVIDIA: CLASHING TITANS IN THE AI MARKET

Muhammad Zulhusni | 15 June, 2023

Both acquisitions are part of AMD’s larger strategy to challenge Nvidia’s dominance in the AI computing space with a broad portfolio of what AMD chair and CEO Lisa Su has called “leadership GPUs, CPUs, and adaptive computing solutions for AI inferencing and training.”

In fact, since last year, Su and other executives have articulated a comprehensive AI strategy that addresses what she said is a “multibillion-dollar growth opportunity across cloud, edge, and an increasingly diverse number of intelligent endpoints.”

The post AMD is taking on Nvidia by acquiring Nod.AI appeared first on Tech Wire Asia.

As we navigate through an era of unprecedented technological advancement, Apple is aiming to once again push the envelope with the delivery of the A17 Pro chip.

The tech behemoth has long been synonymous with innovation, but its recent foray into proprietary silicon has ushered in a transformative new chapter. With the unveiling the  A17 Pro chip, Apple isn’t just enhancing its own devices, but setting new industry benchmarks. But enough glowing hyperbole – let’s take an in-depth look at how the A17 Pro chip promises to redefine the boundaries of performance and efficiency. And how it looks likely to deliver on those promises.

Apple’s in-house silicon technology has been a cornerstone of its strategy for years, epitomizing the company’s belief in a symbiotic relationship between hardware and software. The transition away from Intel was more than just a political shift; it was a commitment to innovation that has culminated in the company’s latest offerings.

What is the A17 Pro that Apple announced?

The recent Apple annual product launch did not disappoint. While the iPhone 15 ships with the familiar A16 Bionic chip—already seen in the iPhone 14 Pro and Pro Max—the real star of the show was the brand-new A17 Pro chip, which will be the heartbeat of the iPhone 15 Pro and Pro Max.

BEFORE YOU BUY THE NEW APPLE IPHONE 15 AND APPLE WATCH, READ THIS

Muhammad Zulhusni | 14 September, 2023

The A17 Pro signals a new frontier for Apple’s engineering teams. It’s the company’s inaugural chip designed using a 3-nanometer manufacturing process.

This is a major technological advance from the 5-nanometer process used for the A16 Bionic. What does this mean in layman’s terms? A smaller manufacturing process allows for a higher density of transistors, effectively packing more computing power into a tinier, more efficient package.

To put the minuteness of this technology into perspective, Apple has revealed that some of the components inside the A17 Pro’s transistors measure a mere 12 silicon atoms in width. This feat of nanotechnology is nothing short of astounding. Moreover, the A17 Pro comes jam-packed with a staggering 19 billion transistors, showing a six-billion-transistor leap from the A16 Bionic, further pushing the boundaries of what’s possible in such a small form factor.

Architecturally speaking, the A17 Pro carries forward the well-balanced blueprint of its forerunner, having two high-performance cores alongside four energy-efficient ones. Although the clock speeds of these cores have yet to be released, Apple assured users that they can expect a 10% speed bump, coupled with improved branch prediction mechanisms and higher execution bandwidths.

Apple is never a company to make modest claims, so it’s unsurprising that it has dubbed the A17 Pro “the world’s fastest single-threaded mobile CPU.” Given that competitors are still struggling to keep pace with the A16 Bionic though, Apple’s assertion doesn’t seem too far-fetched.

Apple introduces the A17 Pro. (Source – X)

Additionally, the efficiency cores in the A17 Pro are hailed by Apple as the “most power-efficient mobile CPU cores,” claiming a performance-per-watt ratio that is a solid three times more efficient than those of rival processors.

The chip’s machine learning capabilities are also quite remarkable. Though the neural engine retains the 16-core structure found in the A16 Bionic, it comes supercharged with a 2x performance increase. This enhancement equates to an almost unimaginable 35 trillion operations per second, elevating features such as voice recognition and advanced photo sorting to new heights.

Graphics and gaming: Apple’s A17 Pro as a game-changer

Regarding graphics, Apple is bringing a revolutionary, pro-class GPU to the table. It’s an elevated six-core design, specifically engineered for optimized performance and energy efficiency. The GPU also incorporates novel rendering techniques and boasts a 20% acceleration over its predecessor.

APPLE JUST RENEWED ITS DEAL WITH QUALCOMM, SIGNALING DELAYS WITH IN-HOUSE CHIPS

Dashveenjit Kaur | 12 September, 2023

This is a significant moment for game developers and those immersed in augmented reality. The A17 Pro’s GPU offers hardware-accelerated ray tracing, a feature that ensures photorealistic lighting and shadows in rendered environments. This hardware-level support allows for a fourfold increase in ray tracing speed compared to software-based solutions, offering higher frame rates and a more immersive user experience.

Apple is also announcing the inclusion of heavyweight gaming titles like Resident Evil 4, Resident Evil Village, and The Division Resurgence. These high-caliber games will be available on the iPhone, a feat made possible largely because of the new chip’s computational prowess.

A head-to-head comparison: A17 Pro vs. Snapdragon 8 Gen 2

But how does the A17 Pro fare against its main competitor, Qualcomm’s Snapdragon 8 Gen 2? On the CPU front, the Snapdragon is manufactured using a 4nm process and features an octa-core design. But its maximum core speed is just 3.2GHz, falling short when compared to the A17 Pro’s blazing 3.7GHz speed.

As for graphics, Qualcomm had an edge last year with its Adreno 740 GPU. But Apple aims to reclaim its crown with a 20% year-over-year performance improvement and hardware-accelerated ray tracing, giving it a significant leg up in the competition.

Can the Snapdragon compete with the Apple A17 Pro chip? (Source – X)

Apple’s new A17 Pro chip sets the bar exceedingly high in the mobile computing space. As we anticipate head-to-head comparisons with Qualcomm’s next-gen offerings, it’s clear that Apple, at least for now, has regained its leading edge in the world of high-performance, high-efficiency mobile technology.

You can find more about this comparison from Beebom.

The post Let’s break down the A17 Pro chip from Apple – and why it matters appeared first on Tech Wire Asia.

Since the Vietnam War, trade and diplomatic relations between the U.S. and Vietnam have not been at their best. For the U.S., doing business with a communist country was historically considered problematic, to say the least. However, things have changed over the last decade, especially with the U.S.–Vietnam Comprehensive Partnership, which was signed a decade ago.

Today, Vietnam is one of the fastest growing economies in Southeast Asia. However, the country is also aware of China’s influence on its trade. Although communist ideology is no longer considered a threat, the U.S. continues to worry about China’s expansion. This concern includes not only China’s territorial claims in the South China Sea, but also its expanding global influence over the world of technology.

INDIA AND VIETNAM BECOMING ATTRACTIVE ALTERNATIVES FOR CHINA PLUS ONE STRATEGY

Aaron Raj | 12 January, 2023

Despite U.S. sanctions on its supply chain, China remains a powerhouse in industries ranging from quantum computing and 5G to semiconductors. At the same time, China’s low-cost technology has become a key factor for American businesses. Currently, most large American businesses, such as Tesla and Apple, manufacture their products in China.

However, with increasing pressure from the U.S. on how these companies can continue to operate in China, many are exploring other countries as alternative locations for their manufacturing plants. Both India and Southeast Asia have emerged as likely destinations for American businesses looking to continue their manufacturing operations without significantly impacting their budgets.

The state of US-Vietnam relations draws interested eyes in China.

Improving US-Vietnam relations

While India has proven to be a worthy competitor to China, its geographic location could result in slightly higher supply chain costs. As a result, more American businesses are considering Vietnam as the ideal location for their production facilities and supply chain management.

According to a report by The Financial Times, the U.S. and Vietnam have agreed to strengthen their ties in response to China’s growing influence. American companies signed deals and partnerships worth billions of dollars during U.S. President Joe Biden’s recent visit to the country.

The President visited Vietnam after the G20 Summit in India. The deals and partnerships signed included  agreements with Boeing, Microsoft and NVIDIA. NVIDIA will be partnering with FPT, Viettel and Vingroup to deploy AI solutions across the cloud, automotive and healthcare industries. Meta and Microsoft have also partnered with Vietnamese companies to roll out AI software across Vietnam to promote digital transformation.

“We’re deepening our cooperation on critical and emerging technologies, particularly around building a more resilient semiconductor supply chain,” said Biden in a joint press conference late on Sunday with Nguyen Phu Trong, General Secretary of the ruling Communist Party of Vietnam.

“We’re expanding our economic partnership, spurring even greater trade and investment between our nations.”

US President Joe Biden holds a press conference in Hanoi on September 10, 2023, on the first day of a visit in Vietnam. Biden travels to Vietnam to deepen cooperation between the two nations, in the face of China’s growing ambitions in the region. (Photo by SAUL LOEB / AFP)

U.S. and Vietnam semiconductor partnership

The White House affirmed the U.S.’s commitment to increasing support for Vietnam in the training and development of a high-tech workforce. Acknowledging Vietnam’s tremendous potential as a major player in the semiconductor industry, both President Biden and General Secretary Phu Trong expressed their energetic support for the rapid development of Vietnam’s semiconductor ecosystem to improve its position in the global semiconductor supply chain.

“Toward this end, the United States and Vietnam announced the launch of semiconductor workforce development initiatives – supported by initial seed funding of US$2 million from the U.S. government, in conjunction with future Vietnamese government and private sector support,” the White House statement said.

Meanwhile, the U.S. State Department announced its partnership with the Vietnamese Government to explore opportunities for growing and diversifying the global semiconductor ecosystem under the International Technology Security and Innovation (ITSI) Fund, established by the CHIPS Act of 2022.

VIETNAM TO START PRODUCING MACBOOK IN 2023 AS APPLE DECOUPLES FROM CHINA

Dashveenjit Kaur | 21 December, 2022

“Vietnam shows promise as a partner in ensuring the semiconductor supply chain is diverse and resilient. Products ranging from vehicles to medical devices increasingly rely on semiconductors as the building blocks of today’s economy. By building on Vietnam’s existing strengths in assembly, testing, and packaging. This collaboration strives to identify new opportunities that attract industry investments and expand the technical workforces in both countries.

The partnership is beginning with a review of Vietnam’s current semiconductor ecosystem, regulatory framework, and workforce and infrastructure needs. The outcome of this review will inform potential future collaboration on developing this critical sector,” said the State Department.

Several U.S. companies also announced plans to build manufacturing plants in the country. Marvell Technology had already announced plans to build a world-class design center in Ho Chi Minh City, Vietnam in May. The site will be home to advanced semiconductor engineering and will be a top technology workplace for employees to enhance their skills and achieve outstanding career opportunities.

Apart from Marvell, Amkor unveiled a new US$1.6 billion factory in the Bac Ninh Province. The plant will assemble, package and test chips, with operations beginning at the end of 2023. Synopsys will also be starting a design and incubation center in conjunction with Saigon Hi-Tech Park.

U.S.-Vietnam relations – a new chapter?

The post Vietnam could be the answer to U.S. chip supply chain appeared first on Tech Wire Asia.

As the semiconductor industry in China continues to grow and become self-sufficient, there is mounting concern in the US. Specifically, the Department of Commerce and Defense in the US is worried about how China’s semiconductor industry growth could impact national security.

Both departments have signed a memorandum to extend collaboration to strengthen the US semiconductor defense industrial bases. To be precise, both departments will increase information sharing to facilitate the coordination of the CHIPS for America’s incentives program. This is also to ensure that respective investments position the US to produce semiconductor chips that are essential for national security and defense programs.

CHIP PLAYERS ARE URGING THE US TO RELOOK INTO CURBS AGAINST CHINA

Dashveenjit Kaur | 25 July, 2023

According to a press statement, Dr. Laura Taylor-Kale, Assistant Secretary of Defense for Industrial Base Policy said the agreement is an important step forward in increasing the capacity and resiliency of the domestic semiconductor industry base.

“It is essential for the Department of Defense (DoD) and Department of Commerce (DoC) to consult one another to ensure we are making complementary investments that support a robust semiconductor industrial base. Both Departments are working together to expand domestic semiconductor production capacity in a coordinated fashion,” said Dr Taylor-Kale.

For CHIPS Program Office Director Michael Schmidt, the agreement will enable teams to coordinate the national security review of applications, produce semiconductor chips in America that the military relies on, and bolster domestic supply chain resiliency.

Chips for America is part of President Biden’s economic plan to grow investment in the semiconductor industry in the US. A total of US$ 52 billion has been allocated to the plans, which also include stimulating private sector investment, the creation of good-paying jobs, and the revitalization of marginalized communities.

Fab equipment spending continues to increase. (Image – Shutterstock)

The state of the global semiconductor industry

The global semiconductor industry faces heavy competition as the US continues to control China’s semiconductor supply chain. China, which is one of the world’s largest semiconductor producers has already had several sanctions on chip imports. However, the semiconductor industry in the country has been working on alternatives within the country to ensure that production continues and that a global supply chain shortage of semiconductors is avoided.

In fact, according to a report by SCMP, sales of semiconductor equipment to China dropped 23% year on year in the first quarter of 2023 amid US export curbs, according to SEMI data. Interestingly, the US and the rest of North America bought 50% more semiconductor equipment during that period.

SEMI, which monitors global semiconductor manufacturing also highlights in its March report that while China is forecast to place third in equipment spending worldwide in 2024, U.S. export controls are expected to limit the region’s spending to US$16 billion, comparable to the region’s investments in 2023.

In 2024, the US is expected to remain the fourth largest region in spending with a record US$11 billion in investments. Meanwhile, Europe and the Middle East are also forecast to log record investments next year, increasing spending by 36% to US$8.2 billion. In Japan and Southeast Asia, fab equipment is expected to increase to US$7.0 billion and US$3.0 billion, respectively, in 2024.

According to a report by Reuters, US Commerce Secretary Gina Raimondo said the Biden administration is seeking to carefully target U.S. controls on exports to China, but rules will cost firms some revenue. The report quoted Raimondo saying “that you deny American companies revenue and China can get the product elsewhere, or China can get the product from other countries.” Raimondo added that rules “will deny some revenue to American companies, but we think it’s worth it.”

US Commerce Secretary Gina Raimondo said the Biden administration is seeking to carefully target U.S. controls on exports to China.

China’s semiconductor industry turning to legacy chips

Meanwhile, the semiconductor industry in China is working round the clock to find solutions to the bans being imposed on them. A report by Bloomberg interestingly highlighted that US and European officials are now growing increasingly concerned about China’s accelerated push into the production of older-generation semiconductors. The concerns have led to officials debating new strategies to contain the country’s expansion.

The report also stated that officials are worried that Chinese semiconductor companies would dump these legacy chips on the global markets in the future. They feel that Western companies may then have no choice but to be dependent on China for semiconductors.

For the Chinese semiconductor industry, apart from the global supply chain disruptions due to the COVID-19 pandemic, no matter how many bans are implemented, they continue to find a way to avoid disruption and continue with their chip production. What makes it more interesting is many countries are still doing business with them, which is why the US feels that they need to be more prepared to deal with any potential national security issues.

As such, it only makes more sense why the DoD and DoC are working together in chip development in the semiconductor industry in the US. The end game is to ensure that the chips are not only made in the US but also go through a national security review and are used by the defense industry.

The post As China’s semiconductor industry grows, concerns mount in the US appeared first on Tech Wire Asia.

As China’s export restrictions loom, the prices of vital chipmaking materials are on an upward trend. In July, the Chinese Ministry of Commerce announced forthcoming controls on the export of gallium, germanium, and some of their compounds. This move requires sellers intending to export these key materials from China to apply and obtain official approval beforehand.

Beijing justifies these impending measures as a ‘safeguard to national security and interests,’ set to take effect on August 1. However, many perceive these actions as a countermeasure to the technology export restrictions that the US and other countries have imposed on China.

According to Nikkei Asia, gallium prices have seen an 18% increase since the end of June, reaching US$332.50 per kilogram in the US and European markets. Meanwhile, the cost of germanium has risen modestly by 4% this month, hitting about US$1,390 per kilogram.

INTEL TO UTILIZE ITS 18A PROCESS TO MAKE 5G CHIPS FOR ERICSSON

Dashveenjit Kaur | 26 July, 2023

Remember, germanium is crucial in high-speed computer chips, specific plastics, and military technologies such as night-vision devices and satellite imagery sensors. Gallium’s applications include the construction of radars, radio communication devices, satellites, and LEDs. This makes these elements critical to the development and performance of many high-tech devices and systems, from consumer electronics to military hardware.

Concerns have arisen that a diminished supply of either material could inflate the manufacturing costs of certain electronic products, or potentially impede the advancement of newer, sophisticated chips. This could have a ripple effect across the technology sector, slowing innovation, increasing consumer costs, and potentially changing the competitive landscape among tech companies.

China: A dominant force in the global supply of key materials

While estimates differ, China is believed to be the world’s principal source of both metals. It’s estimated to account for approximately 60% of the global germanium supply, with the rest primarily sourced from Canada, Finland, Russia, and the United States. Gallium’s global supply, meanwhile, is believed to be at least 80% Chinese, with some estimates as high as 98%.

Germanium is crucial in high-speed computer chips. (Source – Shutterstock)

Reports from July indicated that companies had begun stockpiling these materials or intermediate products that rely on them.

Freiberger Compound Materials, a German company that uses gallium sourced from China to produce its gallium-arsenide semiconductor wafers, informed Reuters of orders for inventory increases as customers rush to stockpile. CEO Michael Harz noted that the industry was quite tense over the situation.

However, as The Register reported following the announcement of the export restrictions, some experts argue there’s no immediate cause for alarm. Neither gallium nor germanium is currently in short supply, and any attempts to restrict their supply could trigger production increases elsewhere.

This viewpoint is reinforced by The Stimson Center, a US-based think tank, asserting that these new controls will push countries to implement policies reducing their reliance on China for vital materials, a process that could take years.

Echoes of the past: Recalling China’s rare earth export restrictions

They reminded that China attempted similar strategies with the so-called rare earth elements in 2010, leading importing countries to diversify their supply sources away from China and sparking an increase in smuggling, causing minimal impact on rare earth shipments from Chinese sources.

Concerns have surfaced over potential restrictions on rare earth exports from China, recalling its shipment curbs from 12 years ago during a dispute with Japan. China is the largest global producer of these metals, integral to EVs and military equipment.

However, China has already issued a foreboding warning that the upcoming export controls could mark the start of their countermeasures in the chip wars, and they could broaden these measures to other crucial materials if external restrictions on China keep escalating.

Following a series of previous curbs, Washington is weighing up additional restrictions on high-tech microchip shipments to China.

The United States and the Netherlands are further anticipated to limit chipmaking equipment sales to China, to prevent China’s military from exploiting their technology.

The day after the announcement of these curbs, Chinese President Xi Jinping reemphasized his call for “stable and smooth functioning of regional industrial and supply chains” during a virtual address to the leaders at the Shanghai Cooperation Organization summit, as reported by state media.

As China flexes its export muscle on critical chipmaking materials, a ripple effect is set to shake the world’s semiconductor industry. China’s strategy is a stern reminder of the importance of diverse sourcing for maintaining resilient supply chains. The immediate surge in gallium and germanium prices is just a prelude to the potential long-term ramifications of such a geopolitical chess move.

However, this is not just an issue for the present, but a harbinger of what the future of global tech dominance could entail. It signals a new front in the technology war, where resource control might become a common weapon. Such scenarios could fundamentally reshape the global tech industry, determining which countries and companies can lead in developing next-generation technologies.

Nations reconsidering supply chain strategies

Nations that previously relied heavily on China’s resources must rethink their supply chain strategies and develop contingencies for such situations. This could lead to a global tech landscape reshaped by the urgency to find alternative resources, potentially opening doors for other countries to enter these roles.

Moreover, while immediate panic may be overstated, the prospect of escalating ‘chip wars’ and broader export controls from China should not be taken lightly. This could set a precedent for similar tactics involving other critical resources, potentially leading to a cascading effect on several technology sectors.

Ultimately, this situation underscores the intricacies and vulnerabilities of global tech supply chains. It’s a wakeup call for industries and governments alike to foster cooperative solutions and diversification strategies to ensure stability in the face of geopolitical turbulence.

The post Chips at stake: China’s export controls and a semiconductor supply chain shakeup appeared first on Tech Wire Asia.