How FuriosaAI Became South Korea’s AI Chip Disruptor

South Korean AI chip startup FuriosaAI turned down a reported $800 million acquisition offer from Meta (NASDAQ:META) in order to remain independent and build its own AI hardware mission‑driven business, TechCrunch reports.

FuriosaAI then secured a major partnership to supply its purpose‑built RNGD, pronounced “Renegade,” chip to LG AI Research’s recently launched EXAONE 4.0 large language model platform, targeting enterprise deployments across electronics, finance, telecommunications, and biotechnology sectors, TechCrunch says.

FuriosaAI declined Meta’s acquisition bid in March due to disagreements over post‑acquisition organizational restructuring and strategic direction, rather than price concerns, according to TechCrunch.

Don’t Miss:

FuriosaAI CEO June Paik, a former Samsung and Advanced Micro Devices (NASDAQ:AMD) engineer, told TechCrunch that the company wanted to continue its mission of sustainable AI innovation on its own path, resisting a buyout that conflicted with its vision and operational autonomy.

TechCrunch says that rejecting a significant acquisition offer signals FuriosaAI’s belief that independence and long‑term control are more valuable than a quick large check from Big Tech.

Founded in 2017, FuriosaAI remains a compact operation with around 15 core staff across Seoul and Santa Clara, California, yet manages global impact through key partnerships and a cutting‑edge hardware design approach, TechCrunch says.

RNGD is FuriosaAI’s second‑generation data center accelerator, built on Taiwan Semiconductor Manufacturing Co.’s (NYSE:TSM) 5-nanometer process and engineered specifically for large language model and multimodal inference workloads, according to the company website.

Trending: $100k+ in investable assets? Match with a fiduciary advisor for free to learn how you can maximize your retirement and save on taxes – no cost, no obligation.

At its core is the tensor contraction processor architecture, which FuriosaAI says treats tensor contraction as a first‑class operation instead of relying on fixed‑size matrix multiplication primitives, unlocking higher efficiency and flexibility in deep learning computations.

Supported by the Furiosa SDK and a full software stack for compression, runtime, and deployment, RNGD enables enterprises to deploy state‑of‑the‑art models with high throughput, low latency, and reduced total cost of ownership.

According to South Korean outlet The Bell, FuriosaAI has recently made strategic hires to bolster its hardware and software leadership ahead of its next-generation AI chip rollout. On July 15, CEO Paik appointed former Korea Advanced Institute of Science and Technology Associate Professor Ji Hun Kang as chief research officer.

According to The Bell, Kang will oversee next‑gen compiler and software architecture design, bringing deep expertise in parallel computing, compiler design, and system verification.

See Also: Warren Buffett once said, “If you don’t find a way to make money while you sleep, you will work until you die.” Here’s how you can earn passive income with just $10.

At the same time, The Bell says that Yeong Jin Cho, a former Samsung executive who led the solid-state drive controller development for 16 years, has joined as vice president of the hardware department. Cho will lead the development of FuriosaAI’s third‑generation AI chip, which is currently targeted for mass production by 2027.

With LG’s EXAONE model now using RNGD servers, FuriosaAI anticipates demand not only from South Korean enterprises but also international customers. According to TechCrunch, the startup is raising about $48 million in funding this month to support scaling and international sales.

With top-tier talent in software and hardware, TechCrunch says FuriosaAI is doubling down on its independent mission to challenge graphics processing unit leaders like Nvidia (NASDAQ:NVDA), accelerate enterprise AI adoption globally, and ensure its RNGD architecture leads in energy efficiency and performance.

Read Next: How do billionaires pay less in income tax than you? Tax deferring is their number one strategy.

Image: Shutterstock

Up Next: Transform your trading with Benzinga Edge’s one-of-a-kind market trade ideas and tools. Click now to access unique insights that can set you ahead in today’s competitive market.

Get the latest stock analysis from Benzinga?

This article From Meta’s $800 Million Offer To LG Deal: How FuriosaAI Became South Korea’s AI Chip Disruptor originally appeared on Benzinga.com