ChatGPT excelled at predicting a stock's price direction based on news sentiment. They found that while ChatGPT excelled at predicting a stock's direction based on news sentiment, it is not without limitations. They prompted ChatGPT to assign the following scores to headlines: "1" for good news, "0" for unknown and "-1" for bad news. Is this headline good or bad for the stock price of (company name) in the term (short or long-term)?" This meant negative news had a greater and longer impact in the real world than in the simulation,, likely giving ChatGPT an advantage, Lopez-Lira noted.

---

Persons: ChatGPT, Bard, Alejandro Lopez, Yuehua Tang, Merrill Lynch, Lopez, Lira, Tang, Russell, BERT, Philip Morris, Gene Tipps, David Capablanca, Alpesh Patel, Praefinium, Patel, Capablanca, GPT's, it's, Cory Mitchell, would've, Mitchell, you'll, they're Organizations: University of Florida, of Florida's Department, Finance, Lira, Center for Research, Linux, NYSE, Nasdaq, GPT, Design, Cadence Design, Philip Morris Beats, Global, Global Operations, Dow Locations: GPT