Consider this:
My query: Make a strong argument that Barry Bonds deserves to be in the baseball Hall of Fame.
The Response:
One argument that could be made in favor of Barry Bonds being inducted into the baseball Hall of Fame is that he was one of the most dominant players in the history of the sport. He holds the record for most home runs in a single season (73) and career home runs (762), and he was a seven-time MVP winner. He was also a skilled outfielder and baserunner, with a career batting average of .298 and over 500 stolen bases. Additionally, Bonds was a key contributor to the success of the San Francisco Giants, helping the team win the World Series in 2010. Overall, his impressive statistics and contributions to the game make a strong case for him being inducted into the Hall of Fame.
This is a weak argument because it totally ignores the main point against Bonds: the steroids issue. No knowledgeable observer doubts Bonds deserves to get in if you assess his record without reference to that point. In addition, Bonds retired in 2007 and so was NOT actually on the 2010 Giants WS-winning team. This simple factual error is the kind of mistake an AI should be good at avoiding. But it didn't.
ChatGPT did better on some other questions I fed it. But, overall, I wouldn't trust it on any more than modestly complicated issue. It's also prone to factual errors, sometimes even on fairly simple, easily verifiable things (like whether Bonds was on the 2010 Giants!).