---
title: The opposing informational effects of AI trading
url: https://www.ml-quant.com/papers/arxiv/2611.18872/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-11-20
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2611.18872
source_url: https://arxiv.org/abs/2611.18872
featured: 2026-11-20
citations: unknown
topic: Trading, Microstructure & Execution
---


# The opposing informational effects of AI trading

Better AI can improve individual forecasts while weakening price informativeness by substituting public-data analysis for private observation, reducing private information aggregated into prices.

- Source: https://arxiv.org/abs/2611.18872
- Identifier: arXiv:2611.18872
- Released: 2026-11-17
- First featured: Quant Letter No. 140 (2026-11-20): https://www.ml-quant.com/issues/2026-11-20/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Trading, Microstructure & Execution
- Authors: Hajime Shimao

## Abstract (arXiv, CC0)

Better artificial intelligence (AI) can improve individual forecasts without making prices better at revealing new information. In a competitive market with resource-constrained forecasting, more capable algorithms can substitute analysis of public records for private observations, reducing the private information aggregated into prices. The same trades can convey sophisticated public-data analysis to investors unable to perform it themselves. We establish conditions under which these effects coexist and the private benefit of further computation falls. Numerical equilibrium calculations and finite-data neural-market experiments exhibit corresponding patterns. Out-of-sample earnings forecasts and controlled language-model tasks show related patterns of information substitution. In the neural experiment, wider adoption increases the predictive value of price for limited observers, reduces it for advanced observers, and lowers the benefit of upgrading by 44\%. Retraining by investors who did not upgrade can amplify the loss of private-information precision. The informational consequences of AI trading depend on what prices reveal and who can use them.

## Related

- [FAST: Efficient Action Tokenization for Vision-Language-Action Models](https://www.ml-quant.com/papers/arxiv/2501.09747/): A new tokenization scheme, Frequency-space Action Sequence Tokenization (FAST), has been proposed for robot actions, facilitating the training of vision-language action policies for complex and high-frequency tasks.
- [Deep Reinforcement Learning for Active High Frequency Trading](https://www.ml-quant.com/papers/arxiv/2101.07107/): A new Deep Reinforcement Learning framework has been developed for high frequency stock trading, showing potential for profitable long-term strategies.
- [Is price-impact decay statistically identifiable by propagator estimation? Diffusive prices force the inferred kernel to be transient and fine-tuned](https://www.ml-quant.com/papers/arxiv/2611.10574/): A no-go theorem proves that propagator estimation cannot identify asymptotic impact decay because diffusive prices force fine-tuned decay artificially.
- [Hard to Fool: A Bounded LLM Research Loop for Quantitative Finance, and What Two Dozen Campaigns Found](https://www.ml-quant.com/papers/arxiv/2611.06607/): An open-source research loop constrains language-model proposals to strict JSON and enforces immutable data splits, discovery gates, and hash-chained logs to prevent false discoveries in backtest-driven strategy development.
- [SwingGuard: Risk-Conditioned Multi-Agent Swing Trading with Dual-Timescale Governance](https://www.ml-quant.com/papers/arxiv/2610.12988/): A dual-timescale multi-agent framework generates trading strategies conditioned on current risk and maintains ongoing position governance between strategy revisions on cryptocurrency, preserving returns after transaction costs.
- [TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructure](https://www.ml-quant.com/papers/arxiv/2602.23784/): A Model for Trade-Flow in Market Microstructure: TradeFM is a new AI model that improves the analysis of market structures by studying billions of trade events in stocks, leading to better simulations of financial returns.
