The era where success in capital markets was defined by who could access and act on information first is ending. Historically, finance was organised around informational scarcity, with every advantage stemming from speed and precision in interpretation.
Today, however, artificial intelligence is reconfiguring these assumptions by making analytical capability abundant, automated, and deeply embedded in decision-making. This is not merely a productivity upgrade but a structural transition that threatens the integrity and stability of the entire financial system.
As highlighted by Mona Naqvi, managing director at the CFA Institute Research & Policy Center, in her framework for structural change, the investment profession must now prepare for a world where intelligence is no longer the bottleneck but the baseline.
The end of the information age in finance
Modern capital markets are undergoing a fundamental shift that mirrors the transition from agrarian to industrial economies. While prior technological revolutions brought incremental efficiency, the AI era is redefining how professional judgement is formed and exercised. This transition differs from previous ones because systems can now ingest vast volumes of unstructured data and perform cognitive tasks once considered uniquely human.
As analytical power scales, the logic of informational scarcity is being replaced by a reality of analytical abundance. Doctrines like the efficient market hypothesis and modern portfolio theory are coming under review as the conditions under which they were conceived begin to shift.
A framework for structural change
To navigate this shift, a new framework is required based on four interacting forces: capability, adoption, substitution, and recomposition.
Capability involves the technical expansion of what is possible, while adoption describes how widely these tools are integrated within firms. Substitution reallocates tasks between humans and machines, and recomposition reflects the cumulative, market-level restructuring resulting from these changes.
These forces are governed by “push” factors like declining compute costs and “pull” factors like regulatory frameworks and professional norms. Early decisions about governance and professional standards will create path dependence, influencing the resilience of the system for years to come.
Four paths to an automated future
The integration of AI could lead to several distinct market states, depending on how these forces combine. In “augmented markets,” AI enhances workflow efficiency without fundamentally altering market architecture, preserving human oversight in allocation decisions.
Conversely, “competitive divergence” creates a landscape where uneven adoption leads to a wide dispersion in performance and cost structures between AI leaders and laggards.
“Platform convergence” sees the industry rely on a small number of shared infrastructure providers, while “model-mediated markets” represent a state where AI systems assume primary responsibility for signals and risk calibration. In these more advanced states, alpha opportunities may narrow as signal generation becomes commoditised.
The hidden danger of cognitive convergence
One of the most significant risks identified is “cognitive convergence”, where institutions align around similar model architectures and data frameworks. This alignment compresses the interpretive diversity that markets rely on for resilience and creates “monoculture risk.”
When large segments of capital, often managing assets well above $1bn or €1bn, depend on overlapping models, feedback loops intensify, potentially amplifying liquidity shocks and momentum.
This synchronisation means that signal diffusion occurs simultaneously, compressing the time between a signal’s identification and a market-wide repricing. Traditional oversight, which focused on the solvency of individual institutions, may be ill-equipped to handle shared analytical infrastructure spanning the entire market.
Judgment becomes the new scarce resource
As analytical production costs decline, the bases for competitive advantage and value creation are shifting from research insight to system architecture and governance credibility. The investment profession must move its focus from information production toward model oversight and fiduciary stewardship.
Fiduciary responsibility, once traced to individual human judgement, must now account for hybrid decision environments where accountability is distributed across models and oversight functions.
The paper argued that professional standards must evolve to ensure that human oversight remains documentable and auditable. Ultimately, as intelligence becomes abundant, the ability to govern, deploy, and constrain it becomes the most valuable and scarce resource in finance.
