Nvidia Unveils Inference Chips, Targets $1T Revenue and 28-City Robotaxi Rollout
At GTC 2026 Nvidia CEO forecast a $1 trillion revenue run rate by 2027 while unveiling new inference chips (including Groq 3 LPU) and partnerships with IBM, HPE, Adobe and Uber, lifting shares 1.65% to $183.19. Collaborations with Cisco extend its Secure AI Factory platform—cutting deployment timelines from months to weeks—while tie-ups with Uber and T-Mobile target robotaxi launches in 28 cities by 2028 and edge-AI smart city applications on 5G networks.
1. GTC 2026 Forecast and Inference Chip Launches
At the March GTC event, Nvidia CEO Jensen Huang unveiled new inference-optimized chips—including the Groq 3 LPU—shifting the company’s emphasis from training to accelerating model deployments. He projected that strong demand for inference will drive Nvidia to a $1 trillion revenue run rate by 2027, supported by recent partnerships with IBM, HPE, Adobe and Uber.
2. Secure AI Factory Expansion Cuts Deployment Times
Nvidia and Cisco announced an enhancement of their Secure AI Factory platform to help enterprises deploy and secure AI workloads across data centers and edge environments. The updated platform embeds advanced security features and reduces deployment timelines from several months to just weeks, enabling faster enterprise adoption of generative AI.
3. Robotaxi and Smart City Edge AI Partnerships
In autonomous mobility, Nvidia joined forces with Uber to power robotaxis in Los Angeles and San Francisco starting in 2027, scaling to 28 cities globally by 2028 using its DRIVE Hyperion platform and Alpamayo AI model. Additionally, a collaboration with T-Mobile and Nokia will pilot Nvidia’s RTX PRO 6000 Blackwell Server Edition over 5G edge networks to support smart city applications like traffic management and industrial safety.