Artificial intelligence (AI) agents have been gaining popularity over the past couple of years. That isn't surprising, as they can significantly enhance enterprise productivity.
AI agents can operate autonomously and make decisions to complete multistep tasks in real time. They can also interact with other applications to complete those tasks.
According to a forecast from market research firm IDC, the use of AI agents could jump by a whopping 10x by 2027. It further suggests that there could be a massive 1,000x increase in agent-related AI inference demands next year. It is easy to see why. Research firm Gartner estimates that 40% of enterprise applications will have AI agents embedded into them by the end of this year, up from just 5% as of September 2025.
But how can you as an investor benefit from this trend? Well, you might consider buying shares of AI pioneer Nvidia (NVDA 0.98%). Its large language models (LLMs) seem to be powering agentic AI applications across multiple companies and industries.
Image source: Nvidia.
Nvidia is right in the middle of the agentic AI revolution
Nvidia has played a pioneering role in the AI megatrend. Its graphics processing units (GPUs) have provided the key processing power that was used to train and deploy the leading AI models, and much of the core software was written on its CUDA platform. What's more, the company offers an open-source, agentic AI toolkit with which customers can create and optimize AI agents.
NASDAQ: NVDA
Key Data Points
The company noted on its November 2025 earnings call that it was seeing a surge in the adoption of AI agents across several industries. Importantly, several companies are leveraging Nvidia's offerings to build agentic AI applications. From Palantir to Anthropic to Salesforce to RBC to Meta Platforms, Nvidia is the driving force behind agentic AI growth.
Looking ahead, the adoption of Nvidia's agentic AI tools should continue to grow as more enterprises embrace this technology. More importantly, agentic AI is set to drive a stupendous increase in AI inference demand. As a May 2025 article in the MIT Technology Review noted, inferencing now drives 80% to 90% of AI computing power usage.
This isn't surprising, as we have already seen that AI agents perform tasks in multiple stages. As a result, they may utilize multiple AI models to complete a complex task. Nvidia is well positioned to capitalize on an agentic AI-fueled spike in inference demand. The company claims that "for highly interactive, deep reasoning agentic AI," its upcoming Vera Rubin accelerators will deliver AI tokens at one-tenth the cost of its Blackwell processors.
That 90% drop in inference costs is likely to spur greater adoption of the Vera Rubin processors this year. Moreover, the value of the agentic AI market is expected to grow 9x between 2024 and 2030, positioning Nvidia for years of solid growth. It's worth noting that the increasing usage of AI to support new applications, such as AI agents, has led to a sharp spike in the rentals of Nvidia's H100 GPUs in 2026, according to Bloomberg.
With the rollout of Vera Rubin processors scheduled for the second half of the year, demand for Nvidia's chips should remain solid, particularly given the terrific drop in inference costs that they can reportedly deliver.
Investors may be underestimating the stock's upside potential
Nvidia stock is up by about 5% far in 2026, well underperforming the more than 19% jump in the PHLX Semiconductor Sector index. However, this underperformance may not last for long. Analysts are forecasting a 65% jump in Nvidia's earnings in its current fiscal year, well above the 15% average increase in the earnings of the S&P 500 companies.
What's more, eight of the analysts covering Nvidia have raised their earnings-per-share targets for fiscal 2027 in the past 30 days. That trend may continue due to catalysts such as agentic AI. As such, it would make sense to buy this AI stock before it steps on the gas again, especially considering that it trades at an attractive 24 times forward earnings right now.
