Intel / video

Can AI Agents Beat Wall Street at Predicting Earnings? - Ashwin Paranjape, Samaya AI

Why it matters

Shows how to build leak-free evaluations for financial agents and reports where frontier models now edge analyst consensus.

Key takeaways · AI-distilled

Ashwin Paranjape argues stock prediction is too noisy to learn from, while earnings prediction has hard numbers, an expert baseline in analyst consensus, and an explanation for every surprise.
Samaya AI's test is a point-in-time environment that blocks any look at data from after the prediction date, so an cannot see the answer it is asked to forecast.
Across more than 450 earnings reports, the latest frontier models start to beat analyst consensus, and Paranjape argues live data matters more than anything else for financial agents.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.