Classification System for AI Agents in Investment Management
In close cooperation with the DePaul University in Chicago, the Panthera Group is pleased to release a new academic paper, providing practitioners with clear guidance on where, why and how to embed AI agents in their decision design. A succinct can be found below.
Executive Summary
AI will not prevent poor investment decisions, but it can help investors become more consistent, disciplined, and evidence-driven, if we understand what kind of intelligence we are bringing into our investment process designs.
The Age of the Agent
Artificial intelligence has become the industry’s favourite intern, tireless, fast, and increasingly opinionated. From risk dashboards to co-pilots embedded in existing solutions, AI agents are increasingly adopted in the daily workflows of portfolio managers, analysts, and compliance officers. Yet most firms could not precisely describe the “intelligence” they have deployed with AI. Is it a decision-support tool, an autonomous research analyst, or a delegated trader? Without a shared taxonomy, we risk both over-trusting and under-using a technology that is already reshaping how capital is allocated.
Our research team, a collaboration between DePaul University and Panthera Solutions, developed the first multi-dimensional classification system for AI agents in investment management. The team consists of Ivana Zilic, PhD Patrick Wierckx, Michiel Kuhn, and Dr. Markus Schuller. Their findings provide practitioners, boards, and regulators with a common language for evaluating agentic systems based on autonomy, function, learning capability, and governance.
If you cannot classify your AI, you cannot govern it – and you certainly cannot scale it.
Why a Taxonomy Matters
The current landscape mirrors the early days of ESG integration: enthusiasm without standardization. Compliance teams struggle with explainability; portfolio managers fear algorithmic crowding; CIOs face incompatible metrics of “AI maturity.” A taxonomy does not constrain innovation — it clarifies intent. It allows firms to articulate what problem the agent is solving, who is accountable, and how model risk is mitigated. Without such clarity, AI adoption remains tactical rather than strategic.
The Classification System: Three Dimensions of Intelligent Integration
Financial markets are complex adaptive systems (Lo, 2009). Our framework reflects that ecology by mapping AI agents along three orthogonal dimensions. Together, these axes form a decision matrix that reveals both capability and constraint:
1. Investment Process: Where in the value chain does the agent operate?
2. Comparative Advantage: Which competitive edge does it enhance: informational, analytical, or behavioral?
3. Complexity Range: Under what degree of uncertainty does it function: from measurable risk to radical ambiguity?
Strategic Implications for Investment Offices
- Map your ecosystem. Catalogue AI agents and plot them within the framework to expose overlaps and blind spots.
- Prioritize comparative advantage. Invest where AI strengthens existing advantages.
- Institutionalize learning loops. Treat each deployment as an adaptive experiment; measure impact on decision quality, not headline efficiency. Firms that design around decision architecture, not algorithms, will compound their advantage.
Beyond Efficiency: Towards Decision Ecology
Neuroscientist Antonio Damasio reminded us that all intelligence strives for homeostasis — balance with its environment. Financial markets, too, must regain equilibrium, between data and judgment, automation and accountability, profit and planetary stability. Augmented intelligence, properly classified and governed, can serve that reintegration. It allows capital allocation to become not only faster but wiser, learning as it allocates.
Conclusion
The investment industry stands at a crossroads. Either it lets algorithms dictate investment behavior, or it designs a decision ecology where humans and machines learn together. Our multi-dimensional classification system offers the scaffolding for that evolution, a bridge between hype and practice, between autonomy and alignment.
