Agentic AI: the next frontier in artificial intelligence for investors?

Key takeaways

1. What is agentic AI? Agentic AI builds on generative AI by enabling autonomous decision-making and task execution in dynamic environments.
2. Why is it important for investors? Agentic AI could unlock significant productivity gains across industries, creating new growth opportunities – but it’s just one of several emerging technologies.
3. How can investors capture potential while mitigating risks? Diversified exposure to companies enabling AI infrastructure and applications offers a balanced way to participate without relying on single stocks.

AI’s economic impact – too big to ignore

Artificial intelligence (AI) is today an unignorable macroeconomic driver. According to BofA Global Research, AI-related capital expenditures contributed up to 1.3 percentage points to US GDP growth in Q2 2025.[1] Put another way, if you take away AI spending US GDP would be essentially flat.

Looking ahead, McKinsey estimates that generative AI alone could add between $2.6 trillion and $4.4 trillion annually to the global economy, boosting labour productivity growth by 0.1% to 0.6% per year through to 2040.[2]

These figures underscore why AI is often compared to transformative technologies such as telecommunications in the 1980s and 1990s. But unlike telecoms, which revolutionised a single sector, AI is permeating every industry – from e-commerce and logistics to financial services and healthcare diagnostics.

AI is not just about incremental improvements; it is about creating economies of scale and redefining operational models. This breadth of application makes AI a structural growth theme rather than a cyclical trend.

What is agentic AI and why does it matter?

While generative AI (GenAI) dominated headlines until recently, today attention is shifting to agentic AI.

Agentic AI refers to systems that go beyond generating content or answering queries. These systems act autonomously in dynamic environments, using large language models (LLMs) as a foundation but adding layers of reasoning, planning and decision-making.[3]

In essence, agentic AI can break down complex tasks into subtasks, execute them and adapt based on feedback – much like a human agent operating within constraints.

Why is this significant? Because productivity gains from AI depend not only on automation but on autonomous orchestration of workflows. Agentic AI could, for example, monitor cybersecurity threats in real time, initiate countermeasures and verify system integrity without constant human intervention.

This capability introduces efficiency and resilience, though it also raises questions about governance and verification. It is important to stress that agentic AI does not replace human judgement. Like all AI systems, it relies on the quality of its inputs. Incorrect or biased data will lead to flawed outcomes – a reminder that human oversight remains essential.

Analysts expect significant growth in the agentic AI market, with Precedence Research estimating the total market size to grow more than 20 fold in the coming decade[4].

Is the future agentic?

It’s also crucial to recognise that agentic AI is one tool among many, and in this rapidly advancing field the technologies that could prove most transformative in the long term remain uncertain.

Other emerging technologies, such as quantum computing, could amplify AI’s potential even further. Quantum computing excels at solving branching problems – those with multiple interdependent variables – at speeds unattainable by classical systems.

This could enable breakthroughs in optimisation, drug discovery and financial modelling, creating synergies with AI that are hard to overstate.

Investing in AI – navigating the unknowns

AI is being deployed at scale, but the precise technologies that will dominate over the next decade remain uncertain. Today’s leaders may not be tomorrow’s winners. This uncertainty makes diversification[5] critical, in our view.

Rather than trying to pick individual winners, a prudent approach is to seek exposure to companies enabling AI across the value chain – from semiconductor manufacturers and cloud infrastructure providers to software platforms and cybersecurity firms.

These enablers form the backbone of AI adoption, and are likely to benefit as AI becomes increasingly ubiquitous regardless of which specific applications prevail.

AI is not a monolithic theme; it spans multiple industries and use cases. A diversified strategy has the potential to allow investors to capture the upside of this revolutionary technology while seeking to mitigate the risks inherent in forecasting technological trajectories.

[1] Source: https://institute.bankofamerica.com/economic-insights/ai-impact-on-economy.html

[2] Source: https://holisticds.com/en/articulo/ai-economy/

[3] Source: https://www.ibm.com/think/topics/agentic-ai

[4] Assumptions, opinions, and estimates are provided for illustrative purposes only. There is no guarantee that any forecasts made will come to pass.
[5] It should be noted that diversification is no guarantee against a loss in a declining market.