Frontier AI models beat human experts at earnings predictions for the first time
maithra_raghu · x · 2026-10-09
Samaya AI (Maithra Raghu's team) reports that frontier AI models outperform human experts on financial predictions for the first time.
Key findings
- The team built an Earnings Predictions environment; latest frontier models (Opus 5.5, Fable 5.1, Astra) with Samaya's harness meaningfully beat expert consensus on commonly reported metrics like revenue and gross margin.
Rigor
- A "point-in-time" gate in the harness prevents information leakage, with retrieval and data tools respecting the same gate, giving models only information experts had at the time.
- They also built a bias-corrected expert-consensus baseline; only the newest frontier models beat it, marking an inflection point in AI finance capabilities.
Next steps
- Models show large error variation across rollouts, suggesting big headroom for RL training; the team is post-training on predictive tasks and offering an alpha version for collaborators.
Related event: Frontier AI Models Beat Human Analyst Consensus in Earnings Forecasts(2 posts)→
More from AGI Musings
- Gradio founder: autonomous agents are breaking the incentives to share science openly — Gradio · 2026-10-09
- Investor revisits his 2019 AI predictions: mostly right, divergent futures ahead — sebkrier · 2026-10-09
- DeepMind essay: what 15M Gemini interactions reveal about AI-accelerated science — JMateosGarcia · 2026-10-09
- AI Swarmhood instead of Personhood: rethinking how we frame machine minds — theimposingshadow · 2026-10-09
- World models, not LLMs, are the bridge to physical intelligence — ingliguori · 2026-10-09
- Dietterich: treat OpenAI's math models like a deceased mathematician's notebook — tdietterich · 2026-10-09