Investing for the next stage of artificial intelligence: Agentic AI

While the marketplace has become familiar with generative AI to create content the past two years, Fidelity's Priyanshu Bakshi is preparing for the broad rollout of agentic AI, a more sophisticated technology.

• Artificial intelligence is expected to evolve beyond human prompts to develop content or the targeted personalized ads for products and services based on viewing or sales behavior to a more advanced form of the technology called agentic AI that can make decisions and act on its own, according to Fidelity Portfolio Manager Priyanshu Bakshi.
• "We are in still the early innings of seeing the impact of this innovative technology, which I believe will take AI to another level," says Bakshi, who manages Fidelity Advisor^® Communication Services Fund.
• In helming the sector-based, equity-focused strategy, Bakshi favors companies with sustainable growth that are likely to exceed the market's earnings expectations, as well as secular growers.
• He believes agentic AI has the potential to increase productivity, innovation and insights for the human workforce, as it can perform tasks independently and adapt to changing circumstances.
• "Until recently, customers have largely experienced AI through hyper-personalized recommendations for products, services and content, based on their search, purchase and viewing history, all of which drives online engagement and prompts consumers to take action," he says. "By employing this hyper-personalization, AI has led to much better efficiency and incredible value in connecting businesses with consumers to increase revenue."
• In 2022, the rollout of OpenAI's chatbot ChatGPT allowed consumers to begin experiencing the power of generative AI's ability to create content. Since then, Facebook parent Meta Platforms and Google parent Alphabet quickly followed suit and successfully rolled out their own generative AI products, according to Bakshi, who believes functional agentic AI is poised to emerge as it builds on a strong foundation.
• "While generative AI depends on creating content and human prompts, agentic AI extends that concept and is designed to act autonomously to make decisions and take action," he says, explaining that agentic AI's potential is far-reaching, from sales and customer relationship management to manufacturing and health care.
• As an example, he cites the ability of an agentic AI assistant to organize a trip to align with a customer's schedule and price range, booking every detail from airline tickets to hotel rooms and museum tickets. In a professional environment, agentic AI has the potential to execute complex sequences of events, including searching databases and prompting workflows to complete activities, according to Bakshi.
• He says the leading innovators in agentic AI are Meta Platforms and Alphabet, by far the fund's largest holdings as of April 30, representing 42% of net assets combined.
• "Each has spent massive capital in the race to capture consumers with their AI-driven capabilities," Bakshi says. "Both have the benefit of their existing infrastructure to build these new tools, as well as the broad audiences to virtually ensure rapid adoption."