Institutional investors using the same AI models are accidentally creating a silent market trap with no escape.
April 24, 2026
Original Paper
The Algorithmic Consensus Hypothesis: Systemic Risk in the Age of Uniform AI Driven Markets
SocArXiv · 53tja_v1
The Takeaway
Epistemic homogeneity occurs when the world's largest banks and funds use overlapping data and similar machine learning architectures to make trades. Even without explicit collusion, these systems begin to think and act exactly alike, removing the diversity that keeps markets stable. Investors assume that a smarter AI gives them an edge, but the widespread adoption of these tools makes the whole system more brittle. This hidden coordination creates massive systemic risk because every major player will try to exit the market at the same millisecond when a shock occurs. The market's invisible hand is being replaced by a single, uniform algorithmic mind.
From the abstract
This paper introduces and develops the Algorithmic Consensus Hypothesis (ACH), which argues that the proliferation of machine learning-driven investment strategies among institutional investors has produced measurable convergence in factor exposures, signal consumption, and portfolio positioning, generating a new and underappreciated class of systemic risk. Unlike traditional contagion channels rooted in balance sheet linkages or funding structures, algorithmic consensus operates through epistem