New AI framework automatically fills gaps in financial influencer advice to create complete trading strategies without distorting their original intent.

Researchers developed KICL, an intent-preserving policy completion framework that converts partial trading advice from financial KOLs into executable strategies using offline reinforcement learning

The system identifies that KOLs intentionally specify what to trade and why, but systematically omit execution details like timing, position size, and duration

Tested on multimodal discourse from YouTube and X (2022-2025), KICL achieved the best returns and Sharpe ratio performance on both platforms while maintaining zero unsupported trading decisions

The approach treats KOL statements as incomplete trading policies and uses machine learning to intelligently complete missing execution parameters while preserving the influencer's original directional intent