During the recent earnings call, Jensen Huang proclaimed that NVIDIA has successfully reached Artificial General Intelligence (AGI) ahead of expectations. He followed up with a bold forecast: in the future, just 40,000 human employees might manage an astounding number of AI agents, potentially ranging from 400,000 to as many as 4 million, all operating continuously.
What are the implications of NVIDIA's pioneering progress in AGI? The company already commands the largest computing power infrastructure worldwide. Achieving AGI would create a formidable barrier of "computing power that feeds back into research and development."
Once AGI facilitates Recursive Self-Improvement (RSI), NVIDIA could cultivate a vast internal workforce of AI scientist agents. This intelligent assembly would drive innovations by automating the design of advanced GPU architectures and even reformulating CUDA code—an immense leap toward autonomous technological advancement.
This self-accelerating capability would solidify NVIDIA’s market monopoly, potentially excluding numerous competitors who would struggle to keep pace. Thus, NVIDIA might dominate not only in AI labor globally but also reap significant profits during humanity's transition to a silicon-based civilization.
Huang emphasized that the debate over the precise definition of AGI is largely irrelevant at this juncture. In the tech landscape, AGI signifies that AI systems can match or exceed human performance in numerous high-value tasks. Over recent years, major tech companies have engaged in fierce competition to be the first to reach AGI.
In response to inquiries about the obsession with AGI by firms like OpenAI, Huang asserted that for many tasks, AGI has already been achieved. He reiterated this belief, which he first shared publicly on the Lex Fridman podcast earlier this year.
Huang challenged the industry's lack of consensus regarding the definition of "intelligence" or metrics for identifying AGI, likening it to racers sprinting toward an undefined finish line. Instead, he emphasized the transformational advancements in AI capabilities.
The latest advancements show that NVIDIA’s Avo architecture completed the ARC-AGI-3 benchmark test for long-horizon autonomous agents with a perfect score. This achievement is significant, given that the ARC test is acknowledged as a top intelligence evaluation, designed by François Chollet, the creator of the Keras framework. It challenges AI systems to demonstrate human-like abstract reasoning with minimal context.
While previous models from OpenAI and Google struggled to achieve even 50% accuracy in this test, NVIDIA’s system excelled, scoring 100% across all 183 levels in various environments purely through autonomous reasoning.
Such a milestone indicates that NVIDIA has separated itself from traditional AI constraints, moving beyond mere "pattern matching" to embodying a more general cognitive ability capable of tackling unfamiliar and complex problems. AI is no longer just reactive; it can autonomously decompose tasks, learn from experiences, and self-improve—all hallmarks of real AGI as per Huang’s perspective.
Further validating the claim of AGI, NVIDIA introduced the ChipStack AI Super Agent in collaboration with Cadence during Computex and GTC Taipei 2026. The system, powered by Codex and Nemotron, reportedly operates at Level 5 autonomy, drastically accelerating verification cycles—from weeks down to mere hours.
On the financial front, NVIDIA’s recent quarterly report painted an extremely favorable picture, exceeding all market expectations. The company reported a remarkable $96.2 billion in revenue for Q2 of the fiscal year 2027, an increase of more than $10 billion from the previous quarter. Revenue from the data center segment alone surged more than double year-on-year to $89 billion.
Key highlights from the quarter included:
- Total Revenue: $96.2 billion (estimated $92.2 billion, up 106% year-on-year) - Data Center Revenue: $89 billion (projected $85.8 billion, up 117% year-on-year) - Net Profit: $59.7 billion (62% net profit margin, an increase of 6% year-on-year) - Gross Margin: 75% (in line with expectations, up 250 basis points) - Adjusted EPS: $2.22 (expected $2.10, up 120% year-on-year)
This revenue translates to approximately $1.06 billion daily. Looking forward, guidance for Q3 suggests revenues could hit $108 billion, marking a historic milestone for NVIDIA as the first company to surpass $100 billion in quarterly revenue, with daily earnings soaring to nearly $1.2 billion.
In a striking shift from last year, where leading AI labs struggled financially, Dylan Patel of SemiAnalysis noted that AI has transformed into an extraordinary revenue generator. The cost of computing power translates into significant returns, enabling NVIDIA and its partners to achieve substantial revenue growth.
NVIDIA is set to join the elite “$100 billion in a quarter” club alongside Amazon, Apple, and Alphabet. CFO Colette Kress stirred excitement by providing a forward-looking revenue projection for fiscal year 2028, forecasting approximately 70% growth—surpassing market expectations of just 44%. This suggests a total revenue nearing $673 billion for that year, positioning NVIDIA as the second-largest U.S. tech firm by revenue after Amazon.
Huang added in a seemingly modest tone that the anticipated 70% growth is conservative, suggesting demand unimpeded by supply chain constraints could approach nearly 100%. This could lead NVIDIA’s annual net profit to near $450 billion, eclipsing the GDP of many nations.
The core evolution in the tech industry is now understood to be the creation of "profit-generating Tokens." These Tokens, which represent basic units of data processed by AI, are becoming valuable assets. In practical applications, AI now autonomously crafts efficient code or generates investment strategies, translating raw output into profit.
With the profound changes in AI application, the key metrics for success have shifted toward efficiency measures, such as the number of Tokens produced per dollar spent and per watt of energy consumed. This shift explains why tech giants are heavily investing in AI without dwelling on traditional AGI definitions.
Ultimately, those who harness the power of computing will steer this new frontier, characterized by rapidly evolving production relationships and positioning AI as the driving force behind the global economy. NVIDIA's advancements exemplify this as it propels toward being the most efficient engine in the digital economic landscape.
