
Jensen Huang may have met his match, and it's not AMD, but a stealthy South Korean challenger
What makes FuriosaAI's chip so special?
Live Events
Custom-built for AI inference : Unlike Nvidia's general-purpose GPUs, RNGD is optimized solely for AI workloads.
: Unlike Nvidia's general-purpose GPUs, RNGD is optimized solely for AI workloads. 5nm process with HBM3 memory : This enables faster performance with lower energy consumption.
: This enables faster performance with lower energy consumption. Massive efficiency gains: LG AI Research found that RNGD delivered 2.25x faster inference per watt compared to traditional GPUs.
LG AI Research just gave FuriosaAI a massive boost
Why did Meta try to buy FuriosaAI?
Is FuriosaAI really a threat to Nvidia?
Feature FuriosaAI RNGD Nvidia H100 Architecture Custom AI inference NPU General-purpose GPU Process TSMC 5nm TSMC 4nm Memory 48GB HBM3 80GB HBM3 TDP ~180W ~700W Inference efficiency 2.25x GPU per watt Industry benchmark Target use case LLM inference Training + inference
What does this mean for the global AI chip market?
FuriosaAI's edge lies in energy efficiency, affordability, and regional backing.
lies in energy efficiency, affordability, and regional backing. Nvidia's strength remains raw performance and broad developer support.
remains raw performance and broad developer support. Meta's pivot toward in-house chips signals a growing trend toward vertical integration in AI.
Could FuriosaAI disrupt Nvidia's AI dominance?
FAQs: Nvidia vs FuriosaAI:
(You can now subscribe to our
(You can now subscribe to our Economic Times WhatsApp channel
For years, Nvidia CEO Jensen Huang has reigned supreme in the AI chip world. Despite fierce competition from AMD and Intel, Nvidia's dominance in training and running large language models has gone largely unchallenged—until now. A stealthy South Korean startup,, is making waves in the global semiconductor scene and may just be the first true threat to Nvidia's AI hardware empire.Founded in 2017 and backed by heavyweights like Samsung Electronics and Naver Corp,has flown under the radar for years. But in 2025, the startup shocked the tech world by turning down an. That rejection alone signaled confidence—but the real attention came when FuriosaAIInstead of selling, FuriosaAI chose to go big. And now, they're emerging as a serious AI accelerator powerhouse—exactly the kind of challenger Nvidia hasn't faced from Asia until now.At the heart of FuriosaAI's rise is its new, a next-gen AI inference accelerator built to handle the kind of large-scale models powering everything from ChatGPT-style tools to enterprise automation.Here's what sets RNGD apart:That kind of power-to-efficiency ratio could be a game-changer—especially for companies scaling AI operations while trying to manage rising power and cooling costs.FuriosaAI's biggest breakthrough yet came when LG AI Research announced it would integrate RNGD chips into its EXAONE platform, the large-scale AI system used for research across biotech, telecom, and finance.For FuriosaAI, this deal is more than a commercial win—it's validation. LG's evaluation didn't just show superior power efficiency. It also demonstrated better cost-performance compared to Nvidia's popular H100 chips, which are the backbone of today's AI data centers.In other words: FuriosaAI just proved it can compete—and maybe outperform—the world's most dominant AI chipmaker.Meta's failed $800 million bid for FuriosaAI speaks volumes. The social media giant, which is investing billions into building its own generative AI systems, clearly saw value in acquiring a company with proprietary AI hardware.But the deal reportedly fell apart not over price—but. FuriosaAI wanted to stay independent and pursue partnerships with global enterprises instead of becoming a Meta-only operation.Ironically, Meta is also, and is reportedly testing it through Taiwan's TSMC. However, the company's recent friction with Nvidia (Meta was notably excluded from Blackwell GPU order announcements) may have accelerated its chip-buying ambitions.Let's be clear: Nvidia is still the king of the hill. The company's H100 and upcoming Blackwell chips are powering almost every major AI deployment, from OpenAI to Amazon to Google.But FuriosaAI represents something different—a, efficiency-optimized challenger built outside of Silicon Valley. And as, especially at scale, companies will start looking beyond Nvidia for alternatives.Clearly, FuriosaAI isn't aiming to replace Nvidia across the board—but it's carving out a crucial niche in, which is where AI applications go from test labs to real-world products.The race is officially heating up. With Meta, Microsoft, Amazon, and other Big Tech players pushing to reduce their reliance on Nvidia, there's enormous demand forThis competitive shake-up is especially notable because South Korea has long been seen as a memory chip superpower—not a leader in AI accelerators. FuriosaAI's rise could reshape the semiconductor narrative, adding a powerful new player to the global AI arms race.Nvidia's Jensen Huang has faced plenty of rivals before—Intel, AMD, Google TPU—but few with the underdog precision and strategic clarity of FuriosaAI. With top-tier backing, an efficient and powerful chip design, and validation from a major global enterprise like LG, this South Korean startup is signaling that it's ready for the big leagues.Whether it's enough to dethrone Nvidia remains to be seen. But one thing is clear: Jensen Huang may have finally met his match—and it's not AMD, but a stealthy South Korean challenger.FuriosaAI is a South Korean chip startup that makes powerful AI chips, and it's now gaining attention for offering faster, more energy-efficient alternatives to Nvidia's AI hardware.FuriosaAI's RNGD chip delivers overthan Nvidia's GPUs, using less power and offering better cost performance for AI workloads.FuriosaAI turned down Meta's offer because it wanted to stay independent and pursue its own vision of working with global partners like LG.LG uses FuriosaAI's chips to run its large language models (LLMs) for research in industries like biotech, telecom, and finance.Yes, especially for companies focused on AI inference tasks. FuriosaAI offers a cost-effective, energy-efficient alternative that's catching global interest.FuriosaAI is backed by major South Korean players like Samsung Electronics and Naver Corp, giving it strong support for global growth.FuriosaAI's chips are built for—the part where models are used in real applications, not just training.Not entirely, but it could carve out a large share of the, especially where power efficiency matters most.Industries like telecom, research, biotech, and finance are already testing or using FuriosaAI chips in real-world AI systems.Because FuriosaAI just signed a big deal with LG after rejecting Meta's offer, proving it's ready to compete with the biggest names in AI.
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Economic Times
42 minutes ago
- Economic Times
Meta names former OpenAI researcher Shengjia Zhao as Chief scientist for superintelligence lab
Meta taps former OpenAI researcher Shengjia Zhao as Chief Scientist of its new Superintelligence AI Lab, signaling a bold move in the AGI race. Meta Platforms has appointed Shengjia Zhao, a former OpenAI researcher and co-author of the original ChatGPT paper, as the Chief Scientist of its new Superintelligence AI group, the company confirmed Friday (July 25). Zhao, who joined Meta in June, was a key contributor to OpenAI's early breakthroughs, including the development of ChatGPT and the company's first reasoning model, known as o1. The model helped set off a new wave of 'chain-of-thought' AI systems adopted by companies such as Google and DeepSeek. Meta CEO Mark Zuckerberg announced Zhao's formal appointment in a post on Threads, calling him 'our lead scientist from day one.' Zuckerberg added: 'Now that our recruiting is going well and our team is coming together, we have decided to formalize his leadership role.' Zhao will report to Alexandr Wang, the former CEO of Scale AI, who joined Meta in June as Chief AI Officer. Wang is leading Meta's efforts in building artificial general intelligence (AGI), AI that can think and reason at or beyond human capability. The move comes amid Meta's aggressive recruitment campaign in the AI sector. Over the past two months, the company has hired more than a dozen researchers from OpenAI, Apple, Google, and Anthropic. That includes two top Apple AI scientists, Tom Gunter and Mark Lee, according to Bloomberg. Meta launched the Superintelligence Lab in June 2025 as part of its renewed focus on developing advanced AI models. The lab is separate from Meta's long-standing AI research group FAIR, which will continue to be led by Yann LeCun, who now reports to Wang. Meta's recent hires have drawn attention for the size of the compensation packages involved. Some reports suggested offers exceeding $100 million, although the company has denied rumors of higher figures, including claims of $300 million current open-source model, LLaMA 4, has not yet matched the capabilities of OpenAI's GPT-4 or Google's Gemini models. The company is expected to release a more advanced model, internally codenamed 'Behemoth,' later this expressed optimism about the lab's future, saying, 'Together we are building an elite, talent-dense team that has the resources and long-term focus to push the frontiers of superintelligence research.'


Time of India
an hour ago
- Time of India
Meta names former OpenAI researcher Shengjia Zhao as Chief scientist for superintelligence lab
Meta Platforms has appointed Shengjia Zhao , a former OpenAI researcher and co-author of the original ChatGPT paper, as the Chief Scientist of its new Superintelligence AI group, the company confirmed Friday (July 25). Zhao, who joined Meta in June, was a key contributor to OpenAI's early breakthroughs, including the development of ChatGPT and the company's first reasoning model, known as o1. The model helped set off a new wave of 'chain-of-thought' AI systems adopted by companies such as Google and DeepSeek. Explore courses from Top Institutes in Please select course: Select a Course Category Digital Marketing MCA Finance PGDM Data Science Project Management Design Thinking Public Policy Healthcare Data Science Operations Management Leadership Management CXO Degree MBA Cybersecurity Others Artificial Intelligence Product Management Technology others Data Analytics healthcare Skills you'll gain: Digital Marketing Strategy Search Engine Optimization (SEO) & Content Marketing Social Media Marketing & Advertising Data Analytics & Measurement Duration: 24 Weeks Indian School of Business Professional Certificate Programme in Digital Marketing Starts on Jun 26, 2024 Get Details Skills you'll gain: Digital Marketing Strategies Customer Journey Mapping Paid Advertising Campaign Management Emerging Technologies in Digital Marketing Duration: 12 Weeks Indian School of Business Digital Marketing and Analytics Starts on May 14, 2024 Get Details Meta CEO Mark Zuckerberg announced Zhao's formal appointment in a post on Threads, calling him 'our lead scientist from day one.' Zuckerberg added: 'Now that our recruiting is going well and our team is coming together, we have decided to formalize his leadership role.' by Taboola by Taboola Sponsored Links Sponsored Links Promoted Links Promoted Links You May Like They Lost Their Money - Learn From Their Lesson Best Solutions | Search ads Undo Zhao will report to Alexandr Wang, the former CEO of Scale AI, who joined Meta in June as Chief AI Officer. Wang is leading Meta's efforts in building artificial general intelligence (AGI), AI that can think and reason at or beyond human capability. Aggressive AI push Live Events The move comes amid Meta's aggressive recruitment campaign in the AI sector. Over the past two months, the company has hired more than a dozen researchers from OpenAI, Apple, Google, and Anthropic. That includes two top Apple AI scientists, Tom Gunter and Mark Lee, according to Bloomberg. Meta launched the Superintelligence Lab in June 2025 as part of its renewed focus on developing advanced AI models. The lab is separate from Meta's long-standing AI research group FAIR, which will continue to be led by Yann LeCun, who now reports to Wang. Talent and tensions Meta's recent hires have drawn attention for the size of the compensation packages involved. Some reports suggested offers exceeding $100 million, although the company has denied rumors of higher figures, including claims of $300 million deals. Meta's current open-source model, LLaMA 4, has not yet matched the capabilities of OpenAI's GPT-4 or Google's Gemini models. The company is expected to release a more advanced model, internally codenamed 'Behemoth,' later this year. Zuckerberg expressed optimism about the lab's future, saying, 'Together we are building an elite, talent-dense team that has the resources and long-term focus to push the frontiers of superintelligence research.'


Economic Times
3 hours ago
- Economic Times
Intel slumps as potential foundry exit deepens investor gloom
Intel shares sank 8% on Friday after the company warned of exiting chip manufacturing if it fails to secure a major customer, a potentially drastic move by the new CEO to cut spending and revive the struggling American icon. Lip-Bu Tan said on Thursday he would further shrink Intel's workforce, halt work on two plants in Europe and slow another in Ohio, binning his ousted predecessor's strategy that relied on building costly facilities to restore its manufacturing edge. The plan for such extreme measures follows a surprise second-quarter adjusted loss and a forecast for a bigger-than-expected loss in the third quarter. The weakening financials pointed to more trouble for Intel after years of mismanagement eroded its PC and datacenter market share and left it with almost no presence in the AI market. The disclosures "revive long-unanswered questions on the chances of success for its foundry business the path forward is if Intel does not develop leading edge manufacturing capability," TD Cowen analyst Joshua Buchalter said. "It's hard to understate the significance of this potential outcome in the context of the history of the semiconductor industry." As part of its new strategy, Intel may reserve the advanced 18A manufacturing process for its products and proceed with its next-generation 14A only if it lands a major external customer commits, Tan told analysts on the post-earnings call. The move could put $100 billion in assets at risk and deepen its dependence on rival TSMC, adding strain to margins already running at about half their historical highs. "Intel Foundry is a big story and currently people are questioning how successful 18A is. A failure in 18A will be a broken story," said Hendi Susanto, portfolio manager at Gabelli Funds. Intel was set to lose nearly $8 billion in market value, if current losses hold. Its current valuation of around $100 billion is less than half of Advanced Micro Devices' more than $260 billion. The stock has lagged far behind rivals this year, rising 12.8% compared with AI darling Nvidia's 30% gain and AMD's 34%. Intel trades at a 12-month forward price-to-earnings ratio of 42.55 versus 33.90 for Nvidia and 32.12 for AMD. Since taking the helm in March, Tan has divested businesses, laid off employees and redirected resources as part of his strategic reset to revive the embattled chipmaker. "There are no more blank checks," he wrote in a memo to employees on Thursday.