Skip to main content

AI Committee

In short

11 AI agents debate investment decisions — mimicking a real investment committee

The AI committee runs multiple specialist AI agents that each analyze a stock from a different angle — fundamentals, technicals, sentiment, risk, macro. They debate and vote, like a real investment committee, to reach a consensus recommendation.

The AI committee uses a Blackboard architecture where 11 specialized agents share findings and debate. The Director agent synthesizes inputs and resolves conflicts. Multi-round debates reduce single-agent bias. Each agent's confidence level is weighted in the final recommendation.

Related concepts

  • Deep Analysis — Deep analysis runs all 11 AI agents on a stock simultaneously: fundamental, technical, sentiment, risk, macro, and more. They debate, cross-check each other's findings, and produce a comprehensive investment report.
  • Credits System — Credits are the platform's tokens for running AI analyses. Deep analysis (running all 11 AI agents) costs more credits than a quick overview. Credits reset monthly with your subscription tier or can be topped up.
  • Strategy Alignment — When you pick an investment strategy (like value investing or momentum), the platform scores each stock on how well it fits that strategy. High alignment score = great match. Low score = doesn't fit your style.
  • Reinforcement Learning — RL is how an AI learns to make decisions by trying things and getting rewards or penalties. Think of training a dog — good moves get treats, bad moves get nothing. The AI keeps adjusting until it finds the best strategy.