A new study reveals that going against the investment advice of financial influencers (finfluencers) on YouTube generated significantly higher returns than the S&P 500. Researchers found that an inverse strategy, effectively betting against the finfluencers’ stock picks, outperformed the S&P 500 by 6.8% annually, although with increased volatility. The study also explored the performance of Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) in analyzing these stock recommendations. The complete research paper and associated dataset are publicly accessible online. Further details and discussion can be found in the original Reddit post: [https://old.reddit.com/r/artificial/comments/1lyaxy2/we_benchmarked_llms_and_mllms_on_stock_picks_from/]