The stock rose by more than 5% in early trading, and investors reacted quickly after Nvidia’s chief executive, Jensen Huang, spoke at the company’s GTC developer conference in California.

“I want to thank Samsung who manufactures the Groq LP30 chip for us and they’re cranking as hard as they can,” he said.

He added that the chips are already in production and will be shipped in the second half of the year.

That single update has changed the mindset around Samsung’s chipmaking business. For years, its foundry division has faced challenges, posting heavy losses as it tried to compete with bigger companies. Now, there are new signs of recovery.

At the same event, Samsung displayed the Nvidia chips built on its 4-nanometre process. It also introduced its latest high-bandwidth memory, aimed at handling growing demand from AI systems.

This points to a good working relationship between the two companies, especially as demand for AI chips continues to grow.

Experts say the deal could help Samsung’s foundry unit move closer to breaking even, though there are still challenges. Demand in the mobile market is still weak, and high memory prices could limit profits in the near term.

Even so, the market response shows good reports. Samsung shares were up about 4.3% at one point during the session, after earlier hitting stronger results. The index also moved higher, but not by as much.

There is more to watch this week. AMD chief executive Lisa Su is expected to visit South Korea, where she will meet Samsung chairman Jay Y. Lee.

Her visit is also expected to include a tour of Samsung’s chip plant in Pyeongtaek and it would be her first trip to the country since becoming CEO in 2014.

The company said the tape-out took place in December 2025 and is the first hardware implementation of Ubitium’s universal processor architecture, designed to replace the many specialised processors used in today’s embedded systems.

Embedded computing has grown into a market valued at about $115 billion. But the systems that rely on it have become highly complex. Vehicles are good examples. 

Early models relied on a single processor, but modern cars can contain more than 200 processors, each tied to its own software stack, tools and supplier.

That level of complexity is spreading beyond cars. Robots, drones and industrial machines now run artificial intelligence workloads at the edge. As a result, managing multiple processors and development environments has become a growing challenge for system designers.

Ubitium says its architecture tackles this problem by combining several computing functions into one processor. The design builds on the open-source RISC‑V instruction set architecture, which already powers billions of chips worldwide.

However, the company has expanded the concept beyond a traditional central processing unit. Its universal processor can run Linux and real-time operating systems at the same time. It also handles signal processing tasks such as radar and audio in real time, while executing neural networks for AI inference at the edge. All of this happens on a single chip without separate accelerators or co-processors.

According to Ubitium, the universal processor is fully compatible with existing RISC-V software.

“This tape-out turns a long-held thesis into silicon,” said Martin Vorbach, chief technology officer at Ubitium. “Embedded workloads have outgrown the architectures the industry relies on today. Consolidation isn’t optional anymore. It’s inevitable.”

Ubitium compares its approach to the transition that occurred in wireless technology when software-defined radio replaced fixed-function hardware. In using one programmable processor instead of many specialised chips, the company believes manufacturers can simplify system design, shorten development time and extend product lifecycles.

Several technology partners worked with the startup on the project. These include Samsung Foundry, Siemens Digital Industries Software and ADTechnology.

“The shift toward software-defined, reconfigurable compute is accelerating. Ubitium’s approach, one universal processor replacing multiple specialised chips, aligns with where we see embedded systems heading. We’re proud to manufacture their first silicon.” Said Taejoong Song, vice president and head of Foundry Technology Planning Team at Samsung Electronics.

Jean‑Marie Brunet, senior vice president for hardware-assisted verification at Siemens Digital Industries Software, said early validation tools played a role in the project.

“Shift-left verification helps teams validate system behaviour earlier by running more realistic workloads ahead of first silicon,” he said. “Ubitium’s use of Siemens’ EDA tools, specifically the Veloce CS hardware-assisted verification and validation system, highlights how early validation can de-risk integration, support design closure, and accelerate time to first silicon.”

At the manufacturing stage, backend design support came from ADTechnology.

“Advanced-node silicon delivery depends on disciplined back-end execution across timing, power, and signoff,” said Jun-Kyu Park, chief executive officer of ADTechnology. “We are pleased to have supported Ubitium throughout the implementation process as it progressed to tape-out on Samsung Foundry’s 8nm process.”

The chip validates several core parts of Ubitium’s architecture, including its Universal Processing Array and an LPDDR5 memory interface. The array allows the processor to shift execution modes at runtime, switching between CPU, digital signal processing, graphics processing and parallel acceleration tasks without needing separate hardware blocks.

The company said the design supports general computing, real-time signal processing and parallel AI inference within a single, homogeneous architecture. It also runs standard RISC-V toolchains and modern software frameworks without requiring proprietary languages or compilers.

Ubitium plans a second tape-out later in 2026 as development continues. Volume production of the processor is currently targeted for 2027.

The company says potential applications include radar systems, multi-sensor processing, voice and audio workloads, computer vision, automotive cockpit platforms and industrial human-machine interfaces.