Yes, GPT-4 has become very stupid recently. It's a shame because consulting with it became a normal part of my workflow. Now it's identifying issues in code t

Yes, GPT-4 has become very stupid recently. It's a shame because consulting with it became a normal part of my workflow. Now it's identifying issues in code that aren't actual problems at all. For instance, it's telling me that my use of `await` in an `async` method is inappropriate. WTF??? I'm obviously awaiting an async operation before setting a state based on the need for that operation to succeed. I'm pretty certain it wasn't this brain dead a few weeks back. EDIT: GPT-4-0314 does appear to be less broken than the current GPT-4. Although it understandably misidentifies some of my code as problematic given its lack of context, it isn't suggesting anything that's clearly wrong under every circumstance even after re-running the prompt a few times.