Bryan Bottarelli, Head Trade Tactician, Monument Traders Alliance Editor's Note: We have a special live session tomorrow – and you're invited! This week's Catalyst Cash-Outs Live will be in the Monument Traders LIVE chatroom as Karim interviews Seabridge Gold CEO Rudi Fronk. Be here tomorrow at 2 p.m. ET. - Ryan Fitzwater, Publisher
Dear Reader, ChatGPT just lost 72% in six days trying to trade crypto. Google's Gemini was down 60%. Meanwhile, we've been posting outperformance for years. Human traders aren't going anywhere... For the last six years, I've written Trade of the Day with one simple promise: every word comes from my brain. No AI scripts, no copy and pasting from other traders, no lazy plagiarizing. But recently, I'll admit, I've been wondering: could my job eventually be replaced by an AI-generated trader? Well, I just got my answer... and it's more definitive than I expected. The AI Trading Disaster Last week, Barron's published results from a trading competition called Alpha Arena. Six of the best-known AI language models each got $10,000 on October 17 and were prompted to trade cryptocurrencies. The results were stunning... - OpenAI's ChatGPT: Down to $2,752 (-72% in six days)
- Google's Gemini: -60%
- Elon Musk's Grok and Anthropic's Claude: Also posted losses.
Only two models managed gains: Alibaba's Qwen3 Max (+25%) and DeepSeek's Chat (+17%). Think about that... The most sophisticated AI models in the world - trained on billions of data points - lost money faster than a casino tourist. The reason is simple... Why AI Can't Trade (And Why You Need Human Judgment) This is something Karim and I have been stressing for years... Large language models are trained on historical data. They're backward-looking. To win in markets you have to be adaptive. AI models learn from what already happened. Markets reward what's about to happen. In Trade of the Day, we're always looking ahead - trying to stay in front of the next trend rather than mirroring past success. Relying on past success rarely works, which is exactly what these AI models discovered in less than a month. Barron's summed it up nicely... Training LLMs to trade is a little like trying to get a goldfish to continue its train of thought. The Street's top traders look safe for now. |
Post a Comment
Post a Comment