Jensen Huang states that agentic systems can be kept safe by allowing only two out of three capabilities (access sensitive info, execute code, communicate externally) at any time, not all three.

Jensen Huang states the odds of 100,000 such AI agents building a company like NVIDIA are 0%.

Jensen Huang states that people in China are currently teaching AI agents to look for jobs, do work, and make money.

Jensen Huang states that OpenClaw did for agentic systems what ChatGPT did for generative systems.

Jensen Huang states that current programmers and software engineers are at the cutting edge of understanding how to communicate with AI agents using natural language to design software.