🎰 AI agents for stock trading show ambiguous results — Bloomberg.



Developer Jake Nesler trained an agent based on Claude with his own trading logic and launched it on a simulated account with $100,000. Over 30 days, the bot yielded about 7% compared to approximately 4.5% for the S&P 500, but with drawdowns up to 22%. The author does not recommend investing real money.

Structural problem: language models default to conservative positions — blue-chip stocks and defensive assets. Traders are forced to manually reconfigure agents for a more risky profile.

Former quant trader Annanay Kapila believes that LLMs are fundamentally unsuitable for short-term price forecasting.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin