NVIDIA CEO Jensen Huang and Cisco CEO Chuck Robbins sat down for a candid 52-minute fireside chat at the Cisco AI Summit 2026. What emerged was a rare, unfiltered look at how AI is fundamentally reshaping computing and business.
The Programming Revolution Link to heading
Huang described the shift from “explicit” to “implicit” programming as computing’s biggest transformation. Instead of writing code line by line, you now describe what you want in natural language and AI builds it. The barrier between idea and execution is collapsing — you no longer need to be a software engineer to create sophisticated applications.
AI Factories: Where Value Lives Link to heading
Huang introduced the concept of “AI factories” that manufacture intelligence at scale, requiring integrated layers: hardware, energy, infrastructure, models, and applications. His key insight: applications are where companies win or lose. Hardware and models are commoditizing — differentiation happens at the application layer. Build AI-powered applications or risk obsolescence.
Forget ROI, Start Experimenting Link to heading
Their advice for enterprises: Don’t begin with ROI calculations. Start with your core mission and experiment freely. Let a thousand flowers bloom with tools like Claude, Gemini, and ChatGPT. AI’s value emerges from unexpected places — the technology moves too fast to predict from spreadsheets. Even NVIDIA’s internal AI projects are “out of control” in the best way.
Abundance Over Optimization Link to heading
AI creates computational abundance, making previously impossible problems solvable. Huang urged an “infinity perspective” — don’t ask “How do we improve 10%?” but “What’s possible with infinite computational resources?” This reframes AI from optimization to expansion. Drug discovery, climate modeling, protein folding — formerly intractable problems are now within reach.
AI won’t replace physical tools but will augment them. The value isn’t replacing the car; it’s the self-driving capability that makes it worth exponentially more.
Domain Experts as Programmers Link to heading
In the AI era, domain expertise trumps coding ability. Chip designers, analysts, and logistics experts can build tools by describing what they need. They become “implicit programmers” using natural language instead of code. This inverts tech’s traditional power structure — if you can articulate what you want, you can build it yourself. Coding becomes “just typing.” The valuable skill is knowing what to build and why.
Beyond Chatbots to Agentic AI Link to heading
Current chatbots are just the beginning. Real transformation happens when AI systems can reason, plan, use tools, and solve novel problems autonomously. This “agentic AI” can design experiments, optimize supply chains, manage projects — moving from parlor trick to genuine problem-solving partner.
Critical security note: Keep sensitive questions local. Don’t send proprietary queries to cloud AI services. Your questions reveal your strategy; your problems reveal your weaknesses.
The Bottom Line Link to heading
Every company must become an AI company — not by adding a “Head of AI” but by fundamentally rebuilding around intelligence. We’re shifting from human-in-the-loop to AI-in-the-loop, where AI continuously improves itself. This isn’t future speculation, it’s happening now.
The companies that embrace AI as creating abundance rather than threatening scarcity will shape computing’s next era. The question isn’t whether to adopt AI, but whether you’re ready to reimagine what’s possible.