The opposing informational effects of AI trading
Hajime Shimao
Abstract
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.
Create a lesson
Related papers
Structural economic change under the digital transformation
Marco Dueñas, Antoine Mandel
Globally promoting legumes by production or use subsidies
Pablo de la Vega, Wolfgang Britz
Persistence and Change in National Production Networks
Haojie Liu, Zihan Lin
AI, Specialization, and the Organization of Knowledge Work: Evidence from Legal Services
Liu Zhang, Mark Esposito, Isabel Yang
Does the AI Reviewer Alleviate or Aggravate the Maintenance Burden? Evidence from GitHub Copilot Reviewer
Feiyang, Xu, Poonacha K. Medappa et al.
Clicks, Carbon, and Risk: Climate Attention, Persistent Financial Risk, and Real-Sector Activity
Giovanni Angelini, Luca De Angelis