Nvidia’s latest earnings make Jensen Huang’s AGI claim hard to brush off, even if the definition is still up for debate.
Nvidia CEO Jensen Huang said AI has already reached artificial general intelligence for “many tasks,” as the chip giant reported record second-quarter revenue of $96.2 billion, up 106% from a year ago.
The remark came as Nvidia pointed to a new phase for AI, where autonomous agents are no longer just answering prompts but doing productive work that can generate revenue for businesses.
Huang Says AI Is Already Doing Useful Work
During Nvidia’s earnings call, Huang was asked about AGI and whether breakthroughs such as recursive self-improvement could further increase demand for Nvidia’s chips.
His answer was blunt.
For many tasks, Huang said, AI has already reached AGI. But he also downplayed the industry’s obsession with the milestone itself, saying the more important point is that AI is now doing useful and productive work.
That framing matters because Nvidia is not selling AGI as a theory. It is selling the hardware and software stack needed to run the agents, models and AI factories behind it.
Nvidia Posts Record Q2 Revenue
Nvidia reported revenue of $96.2 billion for the second quarter ended July 26, 2026, up 18% from the previous quarter and up 106% year-over-year. The company said GAAP earnings per diluted share reached $2.46, while non-GAAP earnings per diluted share came in at $2.22.
The real engine was Nvidia’s data-center business, which brought in $89.0 billion in second-quarter revenue. That was up 18% from the previous quarter and 117% from a year earlier.
Huang said AI has reached an inflection point because it is doing useful work, generating productive and profitable tokens, and turning compute into revenue.
AI Agents Are Becoming the Next Demand Driver
Huang said AI is moving from systems prompted by people to agentic systems that run in the background and work with other agents.
In his view, companies will not simply use a few AI tools. They may eventually run thousands or even millions of agents that operate continuously, improve workflows and handle business tasks without constant human prompting.
That is the part investors are watching. If AI agents need to run all the time, demand for Nvidia’s compute could keep rising well beyond the current wave of chatbot and model-training demand.
Nvidia Forecasts Another Huge Quarter
Nvidia also issued a strong outlook for the third quarter of fiscal 2027, saying it expects revenue of $108.0 billion, plus or minus 2%. The company said that forecast does not assume any data-center compute revenue from China.
That guidance suggests Nvidia is on track to become a roughly $100-billion-per-quarter company, powered mostly by the global buildout of AI infrastructure.
Still, the company’s growth is not unlimited. Nvidia’s earnings materials and call commentary pointed to supply constraints, memory costs and uncertainty around China as factors that could affect future results.
The Bigger Picture
Huang’s AGI comment will get the headline, but Nvidia’s numbers are the bigger story.
The company is arguing that AI has moved past hype and into a stage where businesses are paying for compute because it produces measurable results. Whether critics agree with Huang’s AGI definition or not, Nvidia’s revenue shows that customers are already spending as if AI agents will become a major part of work.
As of this writing, Nvidia has not framed its latest update as a formal AGI announcement.
Huang’s point was more practical: AI is already useful, already profitable for some workloads, and already driving massive demand for Nvidia’s platform.
via: Nvidia Newsroom | MarketWatch
