LG Innotek gears up for imminent robot parts production
Speaking to a group of local reporters following LG Innotek's 49th annual general meeting in Seoul, Moon said, 'We are in discussions with a significant number of companies regarding camera solutions and we are also developing essential components such as hands and joint motors.'
Although initial production volumes this year may not be substantial, Moon projected a tenfold increase from 2027 with significant increases each year from then on.
His remarks echoed his announcement at CES 2025 in Las Vegas in January, where he mentioned that LG Innotek was collaborating with a 'well-known' humanoid robotics firm and promised to disclose further details upon achieving meaningful milestones.
Industry sources suggest that the electronics parts maker is working with more than half of the 14 humanoid robot firms featured in Nvidia CEO Jensen Huang's CES keynote. These companies include Boston Dynamics, Agility Robotics, Figure, Neura Robotics and Sanctuary AI.
Despite potential tariff risks under US President Donald Trump's administration and a temporary slowdown in the electric vehicle market, LG Innotek remains committed to expanding its Mexico operations.
The company's existing Mexican facility manufactures motors, sensors and automotive camera modules. The new plant, currently under construction, will bolster production capacity for camera modules and lighting solutions, crucial for the automotive sector.
Moon clarified that tariffs on LG Innotek's Mexico-produced goods are absorbed by customers rather than the company itself. However, he acknowledged concerns about potential cost transfers, saying, 'We are actively discussing ways to optimize our global production sites to minimize any adverse impact.'
The expanded Mexico plant is set for completion in July, with mass production commencing in October.
Meanwhile, at the annual general meeting, shareholders approved all four agenda items, including financial statement approval, director and audit committee member appointments, and executive compensation limits.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Korea Herald
7 hours ago
- Korea Herald
Hyundai AutoEver earns top global safety rating for entire vehicle software platform
Hyundai AutoEver, the vehicle software arm of Hyundai Motor Group, said Thursday that its automotive software platform, Mobilgene Classic 2.0, has achieved ASIL-D certification — the highest level of automotive safety integrity as defined by the International Organization for Standardization. ASIL, or automotive safety integrity level, is a global standard that evaluates the functional safety of electrical and electronic systems in vehicles. ASIL levels range from A to D, with D representing the most stringent requirements, signifying that the system is extremely unlikely to fail or cause accidents even in the event of malfunction, according to international certification body Det Norske Veritas. Hyundai AutoEver had previously received ASIL-D certification for 8 modules of Mobilgene Classic 2.0 in 2021, and now it has secured certification for all 63 modules after four years of continued development. Each module plays a vital role in functions such as data transmission, diagnostics and hardware communication, requiring its stable performance for the safety and reliability of the overall vehicle software system. The company's separate high-performance software platform, Mobilgene Adaptive, also received ASIL-D certification earlier this year. 'This certification reflects over 20 years of our expertise in in-house automotive software development,' said Hyundai AutoEver's Software Platform Business Division Director Ryu Seok-moon. 'We will continue to enhance our competitiveness in the software-defined vehicle era.'
Korea Herald
10 hours ago
- Korea Herald
SK hynix posts record quarterly profit on AI chip boom
Chipmaker eyes double HBM sales this year, with next-gen chips ready for Nvidia SK hynix, the world's top memory chip-maker by revenue, hit an all-time high quarterly operating profit of 9.2 trillion won ($6.73 billion) in the April-June period, thanks to soaring demand for its high bandwidth memory chips, a core component in AI processing. Reporting the second quarter's earnings, the company said it aims to double HBM sales and shipments this year compared to 2024, as well as increase its capital expenditure beyond its initial plans for rising HBM demand starting in 2026. The chipmaker reported that it has recorded 22.2 trillion won in sales in the second quarter, up 35.4 percent on-year. The operating profit also surged 68.5 percent over the same period, and net profit came in at 6.99 trillion won. Backed by strong earnings, SK hynix' cash holdings rose by 2.7 trillion won on-quarter to 17 trillion won. Its debt ratio stood at 25 percent, with net debt ratio at 6 percent, down 4.1 trillion won from the previous quarter. "We are on track to meet our goal as a full stack AI memory provider, satisfying customers and leading market expansion through the timely launch of products with best-in-class quality and performance required by the AI ecosystem," Song Hyun-jong, president and head of Corporate Center at SK hynix, said. "We will carry out part of the planned investments preemptively this year for a smooth delivery of major products with visible demand for next year, including HBM." The strong performance was driven by the full-scale ramp up of 12-layer HBM3E sales and increased NAND flash shipments across all applications. The company also attributed its recent success to an aggressive investment by global big tech companies into AI. "There are concerns about a potential demand slowdown in the second half (of the year), but major market fluctuations seem unlikely. We plan to focus operations on products with clear demand visibility," Kim Kyu-hyun, head of DRAM Marketing, said. While HBM4 involves significant technical upgrades — including expanded (input/output) for higher bandwidth, design changes for lower power consumption and the application of logic processes to the base die — SK hynix said it is factoring in the increased production costs as much as possible into its pricing strategy. "We aim to establish optimal pricing in collaboration with customers, while maintaining a healthy level of profitability," said Kim Ki-tae, SK hynix vice president and head of HBM sales and marketing. Regarding the latest US export controls possibly weakening the role of its Chinese manufacturing facilities, SK hynix emphasized that its China fabs will remain a core part of its global memory production strategy. "Shortage in legacy products, such as DDR4 and LPDDR4, are emerging across the industry, and we see our China fab playing a key role in meeting that demand," Song said. He also forecast that long-term demand for the legacy products would remain steady, particularly from Chinese smartphone makers. Capacity for older-generation DRAM chips has tightened in the past months as global production shifts toward HBM chips and newer DRAM standards like DDR5 and LPDDR5. Over Nvidia's plan to supply the H20 chips to China, SK hynix said it is ready to respond quickly if customer demand and supply conditions align. The company currently supplies 8-layer HBM3 — the previous-generation product — for the H20. SK hynix also plans to launch its 24Gb GDDR7 in the second half of the year, which is reportedly intended for Nvidia's upcoming RTX Pro AI accelerator targeting the Chinese market. Additionally, the company is focusing on strengthening its core DRAM technology, which underpins HBM. The company said it is developing next-generation technologies such as 3D DRAM and vertical gate architectures, and plans to begin transitioning to its sixth-generation 10-nanometer-class process in the second half of this year, with full-scale adoption scheduled for next year. In the NAND business, the company said it will take a cautious investment approach, prioritizing profitability and market demand, while continuing product development in anticipation of future market recovery. The company plans to expand sales of QLC-based high-capacity eSSDs and strengthen its product portfolio built on 321-layer NAND. Regarding its M15X fab in Cheongju, SK hynix said it plans to begin operations in the fourth quarter of this year, with mass production of next-generation HBM scheduled to start next year.
Korea Herald
14 hours ago
- Korea Herald
US to supercharge AI sales, loosen environmental rules
The Trump administration released a new artificial intelligence blueprint Wednesday that aims to loosen environmental rules and vastly expand AI exports to allies, in a bid to maintain the American edge over China in the critical technology. President Donald Trump marked the plan's release with a speech where he laid out the stakes of the technological arms race with China, calling it a fight that will define the 21st century. "America is the country that started the AI race. And as President of the United States, I'm here today to declare that America is going to win it," Trump said. The plan, which includes some 90 recommendations, calls for the export of US AI software and hardware abroad as well as a crackdown on state laws deemed too restrictive to let it flourish, a marked departure from predecessor Joe Biden's "high fence" approach that limited global access to coveted AI chips. "We also have to have a single federal standard, not 50 different states regulating this industry in the future," Trump said. Michael Kratsios, head of the Office of Science and Technology Policy, told reporters Wednesday the departments of Commerce and State will partner with the industry to "deliver secure full-stack AI export packages, including hardware models, software applications and standards to America's friends and allies around the world." An expansion in exports of a full suite of AI products could benefit AI chip juggernauts Nvidia and AMD as well as AI model giants Alphabet's Google, Microsoft, OpenAI and Facebook parent Meta. Trump signed three executive orders Wednesday that incorporated elements of the action plan, including the loosening of environmental rules, establishing rules for chip exports and seeking to limit political bias in AI technology. Biden feared US adversaries like China could harness AI chips produced by companies like Nvidia and AMD to supercharge its military and harm allies. The former president, who left office in January, imposed a raft of restrictions on US exports of AI chips to China and other countries that it feared could divert the semiconductors to America's top global rival. Trump rescinded Biden's executive order aimed at promoting competition, protecting consumers and ensuring AI was not used for misinformation. He also rescinded Biden's so-called AI diffusion rule, which capped the amount of American AI computing capacity some countries were allowed to obtain via US AI chip imports. "Our edge (in AI) is not something that we can sort of rest on our laurels," Vice President JD Vance said in a separate appearance at the event, which was organized by White House AI and crypto czar David Sacks and his co-hosts on the "All-In" podcast. "If we're regulating ourselves to death and allowing the Chinese to catch up to us, that's not something ... we should blame the Chinese for ..., that is something we should blame our own leaders for, for having stupid policies that allow other countries to catch up with America," Vance said. The AI plan, according to a senior administration official, does not address national security concerns around Nvidia's H20 chip, which powers AI models and was designed to walk right up to the line of prior restrictions on Chinese AI chip access. Trump blocked the export of the H20 to China in April but allowed the company to resume sales earlier this month, sparking rare public criticism from fellow Republicans. The plan also calls for fast-tracking the construction of data centers by loosening environmental regulations and utilizing federal land to expedite development of the projects, including any power supplies. The administration will seek to establish new exclusions for data centers under the National Environmental Policy Act and streamline permits under the Clean Water Act. Trump directed his administration in January to develop the plan. Trump is expected to take additional actions in the upcoming weeks that will help Big Tech secure the vast amounts of electricity it needs to power the energy-guzzling data centers needed for the rapid expansion of AI, Reuters previously reported. US power demand is hitting record highs this year after nearly two decades of stagnation as AI and cloud computing data centers balloon in number and size across the country. The export expansion plans take a page from deals unveiled in May that gave the United Arab Emirates expanded access to advanced artificial intelligence chips from the United States after previously facing restrictions over Washington's concerns that China could access the technology. (Reuters)
