On June 1, 2026, the global investment community turned its attention once again to Jensen Huang, the man who has become the face of the AI hardware revolution. In an extensive interview with Bloomberg’s Haslinda Amin, the Nvidia CEO delivered a bold message: the market has fundamentally misjudged the value of software companies in the AI era. While investors have handsomely rewarded Nvidia for its semiconductors, stocks in the SaaS and enterprise software sectors have faced significant pressure as many wonder when the real profits from AI implementation will materialize.
The Shift from Tools to Digital Agents
Huang argued that the current pessimism regarding software stems from an outdated understanding of what software actually is. Until now, software has been a tool used by a human to increase productivity. In the era of Generative AI, software is transforming into a "digital agent" (AI Agent)—an entity that doesn't just assist in work but executes it autonomously.
"The market is looking at old subscription models and fearing the cannibalization of jobs," Huang stated. "What they fail to see is that the value generated by an autonomous agent is multiple times that of a simple tool. You are no longer selling access to code; you are selling digital labor." This paradigm shift, according to Huang, will lead to a new explosion in revenue for companies that successfully integrate generative AI into their workflows.
The Disconnect Between Silicon and Code
Over the past two years, Nvidia has seen its market capitalization soar to record levels as cloud giants (Microsoft, Google, AWS) heavily purchased Blackwell GPUs. However, Wall Street began to worry whether the customers of these companies—the end-users of software—were seeing a corresponding return on investment (ROI). Huang addressed this skepticism by emphasizing that infrastructure (hardware) always precedes application (software).
- The training phase requires massive hardware investments, which have already largely taken place.
- The inference phase is where software profits will be harvested.
- Enterprises are currently in a stage of "pilot fatigue," but the transition to full production is imminent.
Huang explained that Nvidia is no longer just a chip company but a platform company. The CUDA software and the AI libraries it offers are the connective tissue that will allow SaaS companies to unlock their profitability.
Geopolitics and Global Demand
In the discussion with Haslinda Amin, the issue of global demand, particularly in Asia, was also raised. Despite export restrictions to China, Huang remains optimistic. "AI is the new industrial revolution," he noted. "Every country wants to have its own Sovereign AI infrastructure. This isn't just about hardware; it's about developing domestic software that understands local language, culture, and laws."
This trend toward "national AI" creates a massive new market for software that did not exist three years ago. Governments and local enterprises are investing billions to avoid total dependence on US Big Tech, a move that favors specialized software companies worldwide.
"The market often looks in the rearview mirror to predict the future, but in AI, the future looks like nothing we have seen before. Software is not dying; it is being reborn as the brain of the global economy."
Conclusion: The Investment Opportunity
Huang concluded that those betting against software today are making the same mistake as those who underestimated the internet in the early 2000s because infrastructure was expensive. Nvidia may have built the roads, but software companies are the ones who will drive the vehicles that carry value to consumers. For Huang, the valuation disconnect between hardware and software is a temporary anomaly that will correct as AI Agents become an integral part of daily life.
