Nvidia bets on robots and driverless cars
Hardware buyers and factory suppliers now depend on how Nvidia (NVDA) allocates chips for automation projects.

Nvidia (NVDA) is turning robots and driverless cars into a supply-control story as much as a new-market story. A report published on Nov. 12 on BigGo Finance said Nvidia agreed to supply more than 260,000 GPUs to South Korean recipients including Samsung Electronics, SK Group, Hyundai Motor Group and Naver Cloud.
A newsletter.semianalysis.com report the same day put the balance sheet behind that move, saying Nvidia's supply and capacity commitments jumped from $119B to $279B in a single quarter, primarily for memory. That makes the robotics push harder to separate from the race to lock down parts before customers and rivals can do the same.
The industrial links are no longer theoretical. That BigGo Finance report said Doosan Robotics is pursuing a robot operating system using Nvidia's physical AI technology, while Kakao Mobility signed a Level 4 autonomous-driving cooperation pact. It also said Hyundai Motor Group plans to apply Level 2+ autonomous-driving technology based on Nvidia's platform to production vehicles in the first half of 2028.
The same report said SK Telecom will build large-scale AI factories on Nvidia's platform, while SK Hynix will jointly develop and supply AI memory including HBM. It also said the market sees little chance of the immediate supply chain being disrupted, which suggests the issue is future control rather than a present factory stoppage.
A report published on www.techtimes.com on Oct. 13 said NVIDIA was in advanced discussions to invest an additional $1 billion in Figure AI, whose humanoid machines ran on BMW's factory floor for eleven consecutive months. The report also said neither NVIDIA nor Figure AI had confirmed the investment discussions, a limit that keeps the equity part of the story unsettled.
There is a simpler reading: Nvidia may be seeding customers while memory remains the binding constraint. Newsletter.semianalysis.com reported that Nvidia’s supply and capacity commitments jumped primarily for memory. Critics including Jim Chanos and Michael Burry argue, according to www.techtimes.com, that NVIDIA's equity portfolio spans GPU cloud providers, frontier AI labs and humanoid hardware makers, creating correlated financial exposure that is not visible in any single line of its income statement or investment portfolio alone.
The factory floor still gives Nvidia a route from chips to use cases. A release carried on Morningstar on Oct. 21 said Advantech will demonstrate industrial AI capabilities spanning factory-level intelligence, edge computing and robotics. It said the MIC-742-AT, powered by NVIDIA Jetson Thor, delivers up to 2,070 TFLOPS (FP4) and synchronizes camera and sensor data for real-time robotic perception.
A household exposed to Nvidia's public share price would notice this through whether confirmed automation deals make the company look more like a supplier with durable demand or an investor taking hidden correlated risk. For institutions, SK Hynix and Advantech gain if hardware commitments keep turning into memory and robotic-perception work; the risk shifts back toward Nvidia if reported investments are denied or commitments fall.
The sources name no scheduled date for the next check. The reading strengthens if later public reports show another rise in Nvidia supply and capacity commitments or more confirmed automation agreements, and weakens if commitments fall or named partners deny the reported Figure AI talks.
What would change it. A later public report says commitments fell or NVIDIA, Figure AI or another named partner denies a core automation investment.
Sources
BigGo Finance Nov. 12semianalysis.com Nov. 12techtimes.com Oct. 13Morningstar Oct. 21
16 statements, each tied to its sourceWhat each statement rests onHow ARCANE checks a story