Nvidia (NVDA 1.24%) published its fourth-quarter results after the market closed on Feb. 25 and once again posted sales and earnings that surpassed Wall Street's expectations. It recorded non-GAAP (adjusted) earnings per share (EPS) of $1.62 on sales of $68.13 billion, beating the analysts' average estimates for EPS of $1.53 on revenue of $66.21 billion.
Yet the chipmaker's shares have actually lost ground since that quarterly report, and that trend has had bearish spillover effects for other tech companies. Many software companies have been particularly hard hit in recent weeks, but comments from Nvidia CEO Jensen Huang suggest that the market may be getting the wrong read on the outlook for top software plays.
Image source: Getty Images.
Huang says the age of agentic AI has arrived
Agentic AI refers to artificial intelligence systems that can operate autonomously or with very little human input to complete complex tasks. During Nvidia's investor conference call, Huang said that agentic systems had reached an inflection point, and specifically pointed to achievements made by Anthropic's Claude.
NASDAQ: NVDA
Key Data Points
Software-as-a-service (SaaS) stocks have experienced big valuation pullbacks in recent months as investors have come to foresee the possibility that many of these companies' business models could be disrupted by AI. However, Huang said he thought that the market had "gotten it wrong" when it came to the bearish outlook for software stocks.
That Huang is taking a bullish stance on agentic AI isn't necessarily surprising. Nvidia's graphics processing units (GPUs) provide the largest share of the processing power for training and running today's most advanced artificial intelligence models, and the chipmaker has been the single biggest winner from the rise of AI. On the other hand, he has a uniquely privileged vantage point when it comes to watching the progression of AI technologies, and his recent comments could point to opportunities for investors.
Huang thinks that the idea that agentic AI is going to kill software companies is based on a misunderstanding about how those agents will operate.
There's a perception that many end users and businesses will have AI agents essentially create new software from scratch so they can avoid paying for the software being offered by SaaS players. In Huang's view, it's more likely that agentic AIs will primarily wind up as users of the existing software tools -- thereby increasing overall demand for those software tools.
In short, Huang predicts that AI agents will be tool users more than toolmakers. He also thinks that software companies will be able to use AI agents to further develop and enhance their products.
Since the beginning of the year, the sell-offs of SaaS stocks have resulted in a roughly $1.6 trillion market capitalization decline across the software industry. Big losers in the category have included Salesforce, Workday, ServiceNow, Adobe, and IBM.
If Huang is right, software valuations could be poised to rebound and march higher over the long term.
Alternatively, there is a good chance that new AI tools will have disruptive impacts on some businesses. With that in mind, investors who are going bargain hunting in the SaaS space should focus on versatile players with strong competitive foundations.
