On June 8, 2026, the marble-floored halls of Seoul’s Shilla Hotel buzzed with the energy of a pivotal meeting between two tech titans: Lunit, South Korea’s pioneering medical AI firm, and NVIDIA, the global semiconductor and computing powerhouse. At the heart of this gathering was a shared ambition—to reshape the future of healthcare through sovereign artificial intelligence (AI), where nations, not just corporations, take the reins in developing and operating AI systems tailored to their unique needs and data.
Lunit’s invitation to NVIDIA’s AI ecosystem summit was no small honor. According to DailyPharm and Medical Times, the event brought together leaders from Korea’s top conglomerates—Samsung, SK Hynix, Naver, Hyundai—and select startups, but Lunit stood alone as the sole representative from the medical AI sector. The meeting wasn’t just a networking opportunity; it was a strategic forum to discuss how sovereign AI could transform not only national infrastructure, but also the global medical landscape.
What exactly is sovereign AI? As IT Times and Biospectator report, it’s a concept championed by NVIDIA CEO Jensen Huang, who was present in Seoul for the occasion. Sovereign AI refers to a nation’s ability to independently develop and operate AI using its own infrastructure, data, and talent. In an era where AI is quickly becoming a backbone of critical industries, this approach is seen as a way to secure national interests, protect sensitive data, and foster homegrown innovation.
Lunit’s role in this emerging ecosystem is anything but hypothetical. Over the past decade, the company has quietly built up a formidable track record. Back in 2017, NVIDIA recognized Lunit as one of the world’s top five socially impactful AI startups through its Inception Awards—a nod that marked the beginning of a close partnership between the two firms. Lunit has since leveraged NVIDIA’s GPU technology for AI training and deployment in hospitals around the world, from Europe and the Middle East to Asia and Australia, according to Pinpoint News.
At the summit, Lunit’s Chief Technology Officer, Sungwon Yoo, took center stage. With over ten years of software development experience at global giants like Google and Intel, Yoo brought both technical depth and a global perspective to the table. He led discussions with NVIDIA’s Jensen Huang and other executives, focusing on how to embed medical science-specialized foundation models onto NVIDIA’s cutting-edge platforms. Yoo didn’t shy away from the challenges, either; he emphasized the need for trustworthy AI systems that reflect national medical data and demographic nuances.
“Each country needs reliable medical AI that incorporates its own healthcare data and population characteristics,” Yoo stated, as reported by Medical Times. “We aim to contribute to the global medical AI ecosystem by collaborating with partners worldwide, drawing on our experience from national screening programs and open medical foundation models.”
Central to these efforts is Lunit’s leadership of a 23-member consortium—spanning top Korean universities, hospitals, and companies—tasked with developing a medical science-specialized AI foundation model as part of a government-backed project. The Ministry of Science and ICT has thrown its weight behind the initiative, providing the consortium with 256 of NVIDIA’s latest B200 GPUs, valued at roughly 18.2 billion KRW (about $13 million USD). These GPUs, which boast two to three times the computing power of previous generations, are crucial for processing the massive omics datasets (molecular, protein, genomic, and clinical data) required to train next-generation medical AI models.
The consortium’s first major achievement, the ‘L1’ model, was recently released as open source. According to Biospectator and Medical Times, L1 is optimized for clinical inference and decision support, featuring an efficient architecture that delivers high performance with relatively low computational demands. It’s already outperformed benchmarks like OpenAI’s HealthBench and is designed to be both precise and ethically robust—no small feat in the sensitive world of medical data. Lunit’s strategy is to set a new global standard by making L1 and future models freely available, driving innovation through openness rather than secrecy.
But Lunit’s ambitions don’t stop at technology. The company is actively applying its models in the real world, having secured national and public cancer screening projects in over ten countries. It has also signed memoranda of understanding with leading Korean hospitals, such as National Health Insurance Ilsan Hospital and Severance Hospital, to accelerate the adoption of AI-driven clinical decision support systems. By integrating multi-layered medical knowledge—from molecular biology to clinical practice—into its foundation model, Lunit aims to create an AI “chain of evidence” that links research, diagnosis, and treatment in unprecedented ways.
Looking ahead, Lunit has set its sights on full commercialization by 2028. The goal? To develop agentic systems that not only assist clinicians with guideline-based answers but also support biomedical researchers in hypothesis testing and drug discovery. The company plans to release all its models as open source, positioning South Korea as a leader in global AI sovereignty—a bold move that could shift the balance of power in the medical technology sector.
Of course, not everyone is without reservations. As JK Economy News highlights, some experts worry about the risks of over-reliance on foreign big tech infrastructure, particularly as global supply chains grow more volatile. There’s an ongoing debate about how to strike the right balance between leveraging the strengths of international partners like NVIDIA and safeguarding national technology sovereignty. For Korean startups, the challenge will be to maintain independence and a sustainable business model even as they collaborate with global giants.
Still, the mood at the Shilla Hotel was one of cautious optimism. The collaboration between Lunit and NVIDIA is seen by many as a catalyst for elevating Korean medical AI to the global stage. With the backing of government, industry, and academia, and a clear commitment to open innovation, Lunit is well-positioned to help redefine how AI is developed, deployed, and governed in healthcare—not just in Korea, but around the world.
As the world watches this partnership unfold, one thing is clear: the future of medical AI will be shaped not just by algorithms and hardware, but by the values of trust, openness, and sovereignty that companies like Lunit and NVIDIA are working to embed at the very core of their technologies.