Bloomberg reported on 28 October 2026 that South Korean AI chip startup FuriosaAI quadrupled its valuation to $2.9 billion in a fresh funding round, bolstering its bid to challenge the likes of Nvidia. FuriosaAI confirmed a $620 million Series D at a post-money valuation of $2.9 billion in a post on its own site the same night.
By our arithmetic, South Korea’s National Growth Fund, which Bloomberg and FuriosaAI both put at about $280 million, supplied about 45% of the round. By our arithmetic, the $620 million is about 21% of the post-money valuation, which implies a pre-money valuation of about $2.28 billion.
Business Pill · Vertical Integration
An evergreen Business Engineer explainer on vertical integration, when a company brings more of its supply chain in-house. It explains the general idea only and does not discuss FuriosaAI.
The key insight: By our arithmetic, Korea’s National Growth Fund supplied about 45% of FuriosaAI’s $620 million Series D, and the round equals about 21% of the $2.9 billion post-money valuation.
FuriosaAI’s post lists international investors Cisco Investments, Infinitum, Jericho Capital, NeoFort Holding, QUDA Capital and Worldline Capital. It says Korea’s National Growth Fund committed KRW 370 billion, about $280 million.
The post adds that the Export-Import Bank of Korea, UAMCO and Hanwha Asset Management joined as new investors, and that its three original seed investors, NAVER through NAVER Ventures, DSC Investment and KDB Capital, contributed again.
Bloomberg puts the international investors’ contribution at $110 million. By our arithmetic, that leaves about $230 million, or 37%, from the remaining Korean and returning investors, a split neither source gives directly.
Bloomberg’s word “quadrupled” implies an earlier valuation of roughly $725 million, by our arithmetic. Neither Bloomberg’s opening paragraphs nor FuriosaAI’s post states the earlier figure.
FuriosaAI’s $620 million Series D by investor group: Korea’s National Growth Fund about $280 million (FuriosaAI and Bloomberg), international investors $110 million (Bloomberg), and the remaining $230 million from other investors. The remainder and all percentages are our arithmetic.
What the Chip Is
FuriosaAI says its RNGD inference accelerator is in mass production: “we’ll produce 20,000 cards this year, and we’ve secured the memory and fab capacity to more than double that in 2027.” The post says RNGD reached mass production in January 2026.
The company describes RNGD as “a 180-watt, air-cooled PCIe card”. Eight of them make its NXT RNGD Server, which it lists at 4U, 3 kW, 4 petaFLOPS (FP8) and 384 GB of HBM3.
FuriosaAI claims RNGD “matches RTX PRO 6000 throughput at the same card count and serves twice as many users per kW of power.” That is the company’s claim; the post gives no independent benchmark.
The post names two customers: Samsung SDS, which it says has launched an NPU-as-a-Service offering powered by RNGD, and LG Group companies, which it says are deploying RNGD in on-premises hybrid cloud environments.
What the Money Is For
FuriosaAI says the round is focused on scaling RNGD production and deployments globally and on building its third-generation chip. It says Gen 3 is in development “with Broadcom and other partners” and will use a 2nm, multi-die design with HBM4/4E memory.
“Taking an advanced AI chip from design to volume production is an exceptionally difficult undertaking, and few companies have demonstrated the ability to do so successfully,” said John Yetimoglu, chief investment officer of Infinitum, in FuriosaAI’s post.
As we read FuriosaAI’s post, the company is pitching RNGD on power, not peak speed: a 180-watt, air-cooled card that it says fits the racks and power budgets enterprises already have.
As we read the investor list, the round mixes a Korean state fund, Korean financial institutions and returning seed investors with US, Singaporean and Abu Dhabi investors named by FuriosaAI and Bloomberg.
FuriosaAI’s post says it will make 20,000 RNGD cards this year and has secured memory and fab capacity to more than double that in 2027; it gives no revenue figure.
FuriosaAI, 28 October 2026
“Our RNGD inference accelerator is in mass production: weโll produce 20,000 cards this year, and weโve secured the memory and fab capacity to more than double that in 2027”
Three Implications
FOR FURIOSAAI The company says the money goes to scaling RNGD production and deployments and to a third-generation chip in development with Broadcom and other partners.
FOR KOREA’S CHIP POLICY FuriosaAI says Korea’s National Growth Fund committed KRW 370 billion, about $280 million, the largest single amount either source names.
FOR NVIDIA’S CHALLENGERS Bloomberg frames the round as bolstering FuriosaAI’s bid to challenge the likes of Nvidia; FuriosaAI names Samsung SDS and LG Group companies as customers.
The framework argues: “This entire ecosystem has been shaped by constraints, bottlenecks, chokepoints, and scarcity at every layer.” It describes AI as a layered stack of technologies and infrastructure.
FuriosaAI’s own post points at one of those constraints: it says power is expensive and scarce. The framework does not discuss FuriosaAI.
What Is Not Established
Neither source gives FuriosaAI’s revenue, the price of an RNGD card or the size of the Samsung SDS and LG deployments. Bloomberg’s opening paragraphs, which are what we read, give no breakdown beyond the $280 million and $110 million figures.
The performance comparison with Nvidia’s RTX PRO 6000 is FuriosaAI’s own and is not independently tested in either source.
This Week In The AI Supercycle
Execution Fault Lines, Concentrated Demand, and Geopolitical Firewalls
This week’s issue: the AI build is not short of demand for its parts. It is short of cheap money, and its end revenue rests on a narrow base of large, often venture-funded buyers.
FuriosaAI raised $620 million at a $2.9 billion post-money valuation, a level Bloomberg describes as four times its previous one. By our arithmetic, Korea’s National Growth Fund supplied about 45% of the round, and the company says it will make 20,000 RNGD cards this year.
A note on sourcing. The valuation and round figures come from Bloomberg’s report of 28 October 2026, of which we read the opening paragraphs, and from FuriosaAI’s own post of the same night. Product, production and customer details are FuriosaAI’s statements. Shares and the implied figures are our arithmetic; linked earlier pieces carry their own sources.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
