Attendees can name their agent, define its personality, grant tools, and use it for tasks like calendar management, travel planning, workouts, or coding, with ongoing learning to deliver new findings.

A major focus is agentic AI, offering practical adoption paths, scaling guidance, and real-world use cases for autonomous workflows and AI-enabled business processes.

GTC 2026 features over 700 sessions, 150 researcher posters, 70+ hands-on labs, onsite certification exams, and real-time developer coverage via a March 18 livestream, spanning topics from physical AI to agentic AI and inference.

The event is described as a dynamic convergence of practical AI development, immersive robotics, and forward-looking discussions on AI-enabled industries, with a festival-like atmosphere.

GTC 2026 centers on AI across healthcare, robotics, and autonomous vehicles, with announcements on future computing and Nvidia’s strategic direction.

Rumored software news includes NemoClaw, an open-source enterprise AI agents platform to help businesses build and deploy autonomous multi-step AI agents.

Key speakers include a marquee Jensen Huang keynote, a panel on open models moderated by Huang with leaders from LangChain, A16Z, AI2, Cursor, and Thinking Machines Lab, and a climate/energy AI discussion with Dario Gil and Ian Buck.

The conference features live updates and hands-on experiences, including an on-site build-a-claw event where participants customize and deploy a proactive, always-on AI assistant using OpenClaw, with access to on-site cloud compute or Nvidia hardware like DGX Spark and GeForce laptops.

There is market speculation about Nvidia’s relationship with Groq, including licensed technology and leadership changes, noted by observers.

Analysts expect software news such as NemoClaw, an open-source platform for enterprise AI agents, and a new AI inference chip architecture to extend Nvidia’s lead from training to inference, competing with Google TPU, Amazon Inferentia, Groq, and AMD.

Disney will present a session on Physical AI and humanoid robotics, showing how AI-powered character systems translate to real-world autonomous characters using modular mechatronics, deep reinforcement learning, and GPU-accelerated simulation.

Discussions will cover AI factories, end-to-end data pipelines, software infrastructure, and scalable AI operations within a broader AI infrastructure focus.