Why Deploying Physical AI at Scale Demands Safety at Every Layer
NVIDIA, Monday, September 21st, 2026
NVIDIA argues physical AI safety must span hardware, software, AI behavior and the full deployment lifecycle as autonomous machines scale up.
As physical AI moves from research into large-scale deployment, NVIDIA argues safety must scale alongside it.
ABI Research projects 49 million level 3-5 autonomous vehicles by 2035, while Omdia estimates roughly 60 million industrial robots will be deployed between 2026 and 2035.
Because roads, factories and warehouses are dynamic environments that can't be fully controlled with static barriers, autonomous systems must perceive changing conditions and reach a safe state when something unexpected occurs.
NVIDIA notes AI behavior needs its own assurance beyond traditional functional safety, pointing to emerging standards like ISO/IEC TS 22440, and stresses that safety assurance must continue throughout ongoing software and model updates.