The AI Agents challenge and the game reshaping the Nasdaq

Artificial Intelligence is entering a new phase. After the surge of "horizontal" generative AI—capable of producing text, images, and code—the market is now turning its attention to AI Agents: specialized software components trained on vertical data and designed to rethink specific business and professional processes.

In the Class CNBC Special that aired on February 17, Edoardo Narduzzi, CEO of Mashfrog Group, described this shift as a natural yet strategic evolution: "Once we discovered horizontal generative AI, we are now discovering vertical generative AI. Agents are specialized software, trained on highly vertical data to reengineer specific processes."

This is not just a technological upgrade, but a paradigm shift that opens up a global competition set to redefine the balance within the tech sector.

Who will win the race for AI Agents

The competition is "quite fierce," Narduzzi observes, and it will unfold on two distinct fronts. On one side are the most advanced and specialized language engines in the B2B space, designed to be integrated into corporate processes by third parties. On the other are the major players that possess vast amounts of data to train increasingly sophisticated and multi-specialized agents.

"Those who win will be the ones with the most specialized natural language–based engines in the B2B space, as well as those who have access to a significant volume of data to train these agents horizontally."

In this scenario, companies such as Anthropic and Alphabet appear to be in particularly strong positions. The game is not only about innovation, but about control of the cognitive infrastructures on which future processes will be built.

Investments, Infrastructure, and the "New" Nasdaq

The full deployment of AI’s potential requires massive industrial investments. Narduzzi refers to "components of an industrial transformation that have not yet been fully realized."

From semiconductors to energy, the ecosystem must evolve to sustain the new technological wave. Micron’s announcement of a $20 billion investment in next-generation chip research is just one example of this infrastructure race.

On the market side, the shift is already visible. Investors are gradually moving their focus from traditional software-as-a-service to companies operating in the AI era. "On the Nasdaq, investors are less interested in pre–artificial intelligence software and far more interested in understanding where AI investments—especially in agentic AI—are headed."

2026 could mark a turning point, with a series of announced IPOs—from Anthropic to OpenAI—poised, if successful, to contribute to the "regeneration of the Nasdaq." Part of the capital will shift from pre-AI era tech companies to those native to the artificial intelligence era. A transition that Narduzzi describes as “a rather unique phenomenon in market economics.”

The AI Scare Trade and market anxiety

In recent weeks, markets have shown signs of nervousness, with what has been dubbed the "AI Scare Trade": sell-offs concentrated in sectors perceived as potentially replaceable by artificial intelligence, from software and financial services to commercial real estate and logistics.

For Narduzzi, the reaction is "somewhat excessive," though understandable. “Agentic artificial intelligence is not just around the corner. It will arrive, but it will still take a few years. Today, we are in the early stage."

However, financial markets thrive on anticipation. In a society where a large share of GDP is based on services, the impact of AI—especially in its generative evolution—is already tangible. In investment banks and consulting firms, tasks once assigned to junior staff are now carried out by tools capable of producing reports, analyses, and due diligence through simple prompts.

This is not just about efficiency. It is a shift that affects the very structure of employment, as evidenced by the growing difficulty young American graduates face in accessing traditional entry-level roles.

Between anticipation and transformation

AI is not yet fully mature in its agentic form, but it is already reshaping expectations, capital flows, and labor dynamics. Markets are anticipating a scenario in which artificial intelligence will no longer be just a tool, but a pervasive infrastructure underpinning the services economy.

The real question is not only who will win the technological race, but how companies, investors, and economic systems will manage this transition. If generative AI has shown what is possible, agentic AI promises to intervene directly in decision-making and operational processes.

The balance between enthusiasm, investment, and industrial sustainability will determine not only the future of the Nasdaq, but also that of an economy increasingly built on software that does not merely assist, but acts.