AI investing: Back to school

Agentic AI in asset management

Asset managers are moving beyond generative AI tools toward agentic AI systems that can support workflows, monitor tasks and assist with defined actions.

In this video-led course, Mike Chen, Head of Next Gen Research at Robeco, explores what agentic AI could mean for the investment value chain: from the search for alpha and the research process, to workflow design, market opportunities and governance. Discover where agentic AI meets human judgment, validation and accountability.

Watch the full video

Explore the five themes of agentic AI in asset management

Each theme opens a different perspective on the shift from AI as a tool to agentic AI as part of the asset management operating model.

Theme 1

Agentic AI and alpha

AI has long supported systematic investing through machine learning, natural language processing and pattern recognition. Agentic AI could take this further by helping investment teams screen ideas, monitor signals, test hypotheses and support portfolio review.

But alpha does not come simply from model access. The real differentiator may lie in the workflow around the model: data quality, validation history, portfolio constraints, mandate boundaries and human sign-off.

Theme 2

Agentic AI and research

Agentic AI is changing not only what investors research, but how research itself is carried out. AI-supported workflows can help scan papers, extract key claims, summarize methods and identify what deserves deeper review.

The opportunity is not to replace researchers, but to strengthen the research engine: preserving what was tested, what failed, what worked and what lessons should carry forward.

Theme 3

Agentic AI and workflows

The first wave of generative AI helped individuals draft, summarize, search and code. Agentic AI points to a broader shift: AI embedded into repeatable workflows across research, monitoring, reporting, client requests, operations, compliance and documentation.

The design question is how to make these workflows safe, reviewable and reversible, and how to fit them into existing controls rather than creating a parallel universe.

Theme 4

Agentic AI and the AI economy

AI is changing what markets reward as well as how investors work. The AI hardware rally has become a major driver of performance, particularly through companies linked to semiconductors, memory, servers, digital infrastructure and supply chains.

This is especially relevant for emerging markets, where parts of the technology ecosystem have played an important role in recent performance. The key investment question is whether the opportunity remains concentrated in a few large names, or is broadening into second-order beneficiaries.

Theme 5

Agentic AI and governance

As AI moves from answering questions to supporting workflows, governance becomes more important. Investors need to know where AI is used, what data it touches, what decisions it influences and who remains accountable when something goes wrong.

The practical questions are clear: where did the answer come from, does it hold up if assumptions change, what is the system allowed to do, and who signs off?

Watch the full conversation

In this 18-minute conversation, Mike Chen explores how agentic AI could affect asset management across five themes: alpha, research, workflows, the AI economy and governance. Each theme can be explored independently or as part of the full journey below.

Explore agentic AI in five key asset management themes

In the (near) future, human quant investors may go from being the direct drivers of research and portfolio positioning to supervising AI algorithms as they carry out those tasks Mike Chen
Mike Chen
Head of Next Gen Research

Warning – Fraudulent use of Robeco on websites and social media Read more