VAST Data, the AI Operating System company, announced on October 13, 2026 that Hyundai Motor Company and Kia, known collectively as HKMC, have made a strategic investment in the company. The investment deepens a relationship that already includes the development of HKMC's Advanced Driver Assistance Systems and robotics AI capabilities on the VAST AI Operating System. The announcement was made in Seoul, South Korea, and follows years of production experience between the two organizations.
Production Foundation for the Investment
HKMC is not new to VAST. Its autonomous mobility division and global software center, 42dot, has been running ADAS research and autonomous driving workloads on the VAST AI OS platform in production. Before standardizing on VAST, HKMC's ADAS and large language model programs sat on disparate platforms from different vendors, fragmenting data and duplicating effort as both programs scaled.
Consolidation on a Unified AI Platform
42dot consolidated these programs into an NVIDIA DGX SuperPOD powered by NVIDIA Hopper built on VAST, bringing ADAS and large language model development onto one data platform. The company is now standardizing its AI data structure and building a Data Flywheel system to accelerate model training and deployment processes. This consolidation is the foundation on which the new investment is built.
Operational Results and Efficiency Gains
In production, the environment has delivered a 2:1 data reduction rate against real sensor and video data. This efficiency offers the footprint and power savings that allow 42dot to expand capacity inside existing data center constraints. VAST's multiprotocol support, including S3 and NFS, lets simulation, training, and validation pipelines work against the same namespace without copies, and the platform replicates 42dot's AI datasets across its data center sites.
Data Demands of Autonomous Vehicles
Autonomous vehicle software is now one of the most data-intensive workloads in enterprise AI. A single sensor-equipped test vehicle generates terabytes of data per day across cameras, LiDAR, radar, and ultrasonics. Multiplied across a global fleet, that becomes hundreds of petabytes under management, with raw sensor data, validation workloads, model updates, and real-time telematics each demanding different throughput, latency, and governance characteristics.
Architecture Built for GPU-Fueled AI
VAST's Disaggregated Shared-Everything architecture, known as DASE, is designed specifically to keep GPU fleets fed at that scale without the data fragmentation and I/O bottlenecks common in legacy high-performance computing file systems. It supports the diverse demands of raw sensor data, randomized video clips, validation workloads, model update syncs, and real-time telematics. This capability is central to HKMC's ability to move quickly from data collection to validated models.
Leadership Perspective
Executive Vice President Hokeun Chung of HKMC's Future Strategy Division said the journey with VAST started as a customer. He explained that consolidating ADAS and language model development onto one platform changed how quickly teams can move from data collection to a validated model. He added that the investment reflects confidence, earned in production, that VAST is the right long-term partner for that scale.
Investor Momentum and Market Position
Renen Hallak, Founder and CEO of VAST Data, said physical AI builders are running into the same data problem enterprises hit years ago, but at a much larger scale. VAST Data closed its Series F financing at a US$30 billion valuation in April 2026, with participation from Fidelity Management & Research Company, NEA, and NVIDIA. HKMC's investment adds to a growing base of strategic and financial investors backing VAST's position as the data foundation for AI infrastructure.
The strategic investment by HKMC signals that VAST's data infrastructure is becoming central to the next generation of physical AI, including autonomous vehicles and robotics. It also shows that production-proven performance can turn a customer into a long-term investor, a distinction that carries weight in enterprise infrastructure. As autonomous driving and robotics datasets continue to grow, VAST is positioned to support the scale, efficiency, and reliability these programs require.