Did you read the article? It's a fairly balanced evaluation of the strengths and weaknesses of LLMs in general, and ChatGPT in particular.