From humanoid workers to industrial manipulators, GTC 2026 showed that AI is finally getting a body.

Hyperscalers still dominate the conversation, but at GTC 2026, the real momentum was with the specialized AI clouds.

What a whirlwind! After four days in San Jose and nearly 400 conversations with exhibitors, from the "Big 8" Elite sponsors to the smallest robotics startups, one thing is clear: we have moved past the "Proof of Concept" phase. AI is now an industrial process—an AI Factory.

In this series, we'll break down everything we learned, sector by sector.

The monetization strategy for private AI is evolving. Do we focus on hosting secure re-training servers, or is the value in providing 'Private Models' as a service? Alternatively, perhaps a data-privacy-first exchange allows for anonymized datasets to be contributed back to the collective model in exchange for reduced costs.

Lots of new information to assimilate, this first-of-many post focuses on private agency related thoughts including putting the puzzle together for Private Agency for My House.

Soumith Chintala (co-creator of PyTorch) provided insights into the evolution of local inference and distributed training, which are foundational for home-based ML.

Ingredients to the bake Private Agency in the home?​

Ignorance is dangerous, so let's take a look at what are the known-knowns, known-unknowns, and unknown-unknowns.

The AI Engineering Summit was a definite eye-opener to the speed with which IT is transforming mundane "busy" work and lowering the startup cost for exploring new ideas.

Take aways for a private agent:​

• Do Agents need to be local to be private?
• Locally on a MacBook Pro (using frameworks like Ollama or llama.cpp)
• Hosted in a secure enclave on AWS / Azure / GCP using Trusted Execution Environments (TEEs)—hardware-isolated areas of a processor that ensure data and code are protected from the host operating system or cloud provider during computation.
• Or can some sort of formal proof be done to leave a multi-tenant agent in the cloud with data privacy? (e.g., exploring Zero-Knowledge Proofs or Fully Homomorphic Encryption)
• For either of the above, how to manage the balance of cost?
• How are we protecting state?
• Where are we storing state?