Countering the narrative that AI will kill software, NVIDIA CEO Jensen Huang argues agents will be tool users, not tool builders. Just as a robot would pick up a screwdriver instead of reinventing one, AI agents will leverage existing platforms. This positions AI as an accelerator for current software, not an immediate replacement.

The dominant paradigm of interacting with computers through graphical user interfaces (GUIs) is temporary. The future is a single, conversational AI agent that acts as an operating system, managing all your data and executing commands directly, thereby making applications and their visual interfaces redundant.

Microsoft's CEO Satya Nadella is promoting a new computing paradigm, borrowed from Notion's CEO. Instead of a tool ("bicycle for the mind"), the computer is now an orchestration layer for vast AI capabilities.

NVIDIA is strategically repositioning itself beyond just hardware. Through collaborations like the one with Groq for inference-specific chips and partnerships with cloud providers, the company is building a comprehensive AI platform that covers the entire AI lifecycle, from training and inference to agent orchestration, signaling a major strategic shift.

While GPUs are key for model training, the next AI wave of autonomous agents relies more on CPUs. The task of controlling and orchestrating multiple agents and tool calls is fundamentally a CPU-based process. This is creating a new hardware bottleneck and shifting focus to CPU manufacturers.

The current AI boom focuses on GPUs for "thinking" (Gen AI). The next phase, "Agentic AI" for "doing," will rely heavily on CPUs for task orchestration and memory for context, creating new investment opportunities in this previously overshadowed hardware.

The evolution from simple voice assistants to 'omni intelligence' marks a critical shift where AI not only understands commands but can also take direct action through connected software and hardware. This capability, seen in new smart home and automotive applications, will embed intelligent automation into our physical environments.

The era of dual-purpose AI chips is ending. The overwhelming demand for real-time processing from AI agents is forcing companies like Google and NVIDIA to create dedicated, inference-optimized hardware. This marks a fundamental and permanent split in the AI infrastructure market, separating training from inference.

Nvidia is challenging Intel and Qualcomm in the PC market with its N1X chip. Instead of just a CPU, it offers a full system (RTX Spark) combining a CPU, GPU, and memory. This integrated approach is designed to optimize PCs for running advanced AI features locally, targeting developers and high-performance users.