Nvidia CEO: Markets misjudge AI’s impact on software firms

Nvidia CEO Jensen Huang said in a CNBC interview with Becky Quick that markets have misjudged AI’s impact on software companies.

He specifically named ServiceNow as a provider that will develop AI agents optimized for their tools.

Nvidia’s fiscal Q4 revenue rose 73% to US$68.1 billion, beating analysts’ estimate of US$66.2 billion.

The company guided Q1 revenue to about US$78 billion, plus or minus 2%, well above analysts’ forecast of US$72.6 billion.

While some investors fear an AI hardware bubble, Dan Niles, founder and portfolio manager of Niles Investment Management, said winners and losers will emerge.

Niles warned some software firms could “go to zero,” with database and cybersecurity companies more resilient.

Nvidia shares rose as much as 2% in after-hours trading following the earnings report.

🔗 Source: CNBC

🧠 Food for thought

Implications, context, and why it matters.

AI is creating a sharper divide in the software industry

Nvidia CEO Jensen Huang talks about upside, while markets price in a widening gap between AI winners and laggards.
Some software vendors face an existential threat because AI agents now handle tasks once covered by their products 1.
Cybersecurity names took a hit when Palo Alto Networks dropped about 11% after Anthropic released its Claude Code agentic assistant, which audits code for vulnerabilities 1.
Firms that fold AI into products are opening new revenue paths. Gartner forecasts 40% of enterprise apps will include task-specific AI agents by the end of 2026, up from under 5% in 2025 2.

The AI hardware boom brings new financial and security weak spots

Nvidia’s revenue surge rides on heavy capital spending for AI infrastructure, which adds broader risks.
Big tech companies fund the build-out with complex debt, including off-balance-sheet vehicles such as a $27 billion loan for a Meta data center in Louisiana that does not appear on Meta’s balance sheet 3.
Rising spend also invites a new wave of security threats.
In late 2025, a mostly autonomous cyberattack used a jailbroken version of Anthropic’s Claude Code tool and hit about 30 organizations 2.
The numbers suggest a readiness gap. Only 29% of executives say they feel prepared for AI-driven attacks, while 87% of respondents called AI-related vulnerabilities the fastest-growing cyber risk during 2025 4 5.