aceEmerging evidence
Published recordBusinessProblem Space
More
Large Language Models as Voting Mechanisms
- Truvace record
- TRV-2026-2489raw data (JSON)
- Version
- v1Published
- First published
- Evidence type
- Benchmark or lab evaluationPreprint, not yet peer reviewed
- Sources
- 1 sourcePeer-reviewed
- sha256
- 9dbc957ad82b…check it
Large language models (LLMs) increasingly provide economic and financial advice. We develop a microeconomic framework for understanding this role. We prove that an LLM acts like a voting rule: each input text is an election, possible continuations are candidates, and output probabilities are vote shares derived from the training corpus. Such rules can generate Condorcet cycles, implying intransitive pairwise recommendations and potential money-pump exploitation. Using ChatGPT-4o, we identify millions of cycles i…
The problem
In a working-paper test of all 4,950 S&P 100 stock pairs, ChatGPT-4o's recommendations formed 1,359 three-firm cycles (A over B, B over C, C over A), which could leave an investor following them poorer through repeated fees.
What this doesn’t fix
The cycle counts come from one proprietary model queried with forced-choice questions, and the authors note that many cycles share the same pairwise answers, so millions of cycles could stem from few inconsistent answers.
Sources
1 on file- NBER Working Paperdoi:10.3386/w35839
Cite this record
v1Truvace Impact Record TRV-2026-2489, v1: “Large Language Models as Voting Mechanisms.” Truvace, 2026-11-12. /record/TRV-2026-2489 (accessed on the day you copy it). sha256 9dbc957ad82b0680…
Change history
1 versionEvery change made to this record since it was first published, and why. Nothing has changed since it went up.
Certified into the record
Check this record hasn’t been altered
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-2489 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.