BREAKING
Nvidia Brings AV Safety to Humanoids
Dual-System Robot Cognition
1
System 1 reflexes
↓
2
System 2 reasoning
↓
3
Split-second action
0
Hz
VLM reasoning
0
Hz
motion output
0
cm
height
0
DoF
total freedom
0
kg
rated payload
Adoption vs Open Challenges
Where it's used
●
Ai2, ETH Zurich
●
Stanford, UC San Diego
●
Cuts integration overhead
Open challenges
●
Certifiable safety
●
Handling rare events
●
Building human trust
Physical AI Needs Trust First
AI NEWS BLITZ
Nvidia is adapting its self-driving safety tech to make humanoid robots safer around people.