The two companies that together supply most of the world's AI computing horsepower are sending a consistent message: the chatbot moment is over, and the era of agentic AI — software that can reason, plan, and act autonomously — is just beginning.
On the hardware side, Arm unveiled a homegrown, AI-tuned server CPU, marking what The Next Platform described as the chip designer "coming full circle" after years of licensing its architecture to others rather than building its own data-center silicon.
Nvidia, meanwhile, laid out what it calls an "open agentic AI world," arguing that the infrastructure required to run autonomous AI agents will demand a fundamentally different computing stack than the one built for large language model chatbots. According to The Next Platform, Nvidia is positioning a new initiative called OpenClaw as a foundational platform for agentic AI — comparing its potential industry impact to what GPT represented for conversational AI.
Nvidia executives also detailed a roadmap for AI systems that goes well beyond current GPU clusters, suggesting the company sees sustained hardware demand as AI applications shift from answering questions to completing multi-step tasks with minimal human oversight.
The Yahoo Finance report noted that executives from both Arm and Nvidia explicitly highlighted this shift toward agentic computing as a defining near-term trend for the industry.
Why it matters: if both the dominant chip architecture licensor and the dominant AI accelerator maker are reorienting their roadmaps around agentic AI, it signals that the next wave of enterprise AI spending — and the hardware it runs on — will look very different from the chatbot infrastructure built over the last three years.