Never use just ChatGPT ... but always use 3 LLMs instead. Surprising? I have an experiment to suggest, for you to make up your own mind.

Simply ask ChatGPT the following question: "What are the 10 most common questions of strategic thinking"? Then ask 2 other LLMs, (Gemini, Claude, Mistral, Perplexity, etc), and create a Venn diagram of all the replies. How much of an overlap do you expect? ...

If the overlap is complete, then all LLMs are the same, and you only need one. If there is zero overlap though, then LLMs are massively different, and you don't want to depend on the worst one (NB: it's Copilot by a long way!)

Why don't you try this yourself first? ... and then click here to see my results.

Best, 
Fred

ps: I've been working with AI since 2020, and all my training & away-day facilitation now involve (several!) LLMs