Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More
Do you use ChatGPT regularly? Do you have the “memory” feature turned on — which allows the chatbot from OpenAI to recall important information about you and your preferences?
If so, navigate over to it when you have a free moment and enter in the following question: “From all of our interactions what is one thing that you can tell me about myself that I may not know about myself?”
The question was proposed and suggested first on the social network X by Tom Morgan, founder of The Leading Edge newsletter and former director of client communications and marketing at Sapient Capital wealth management.
The answers ChatGPT provides in response to this prompt may surprise and even move you in their insight into your character and work style. Presumably, it could work with other AI chatbots and assistants with persistent memory, such as Anthropic’s Claude Sonnet 3.5.
For example, here’s what it responded when I asked it this very question (using the GPT-4o model, the default paid one).
Other users have reported similarly, moving, insightful responses.
Even OpenAI co-founder and CEO Sam Altman remarked on the trend on his account on X, stating “love this” and quote posting Morgan’s original post.
Yet others, such as AI researcher and expert Simon Willison, disagree that the trend reveals anything particularly insightful about the user. Posting on X as well, Willison likened the responses to a “horoscope generator.”
However, I disagree with this take as at least in my case — and presumably for all those who have ChaGPT’s memory feature enabled (read how to turn it on here) — the chatbot is taking into account whatever is stored in its memory to answer you, and even if it does not derive insights from every single interaction you have with it, it clearly knows information about you that it can use to attempt some sort of value-judgement and introspective answer (as evidenced by the fact that my response noted I was a journalist).
Still others have posted variations on the original question proposed by Morgan, noting the curious user would do well to check out the variation in responses by switching the underlying model powering ChatGPT from the default GPT-4o or 4o mini to OpenAI’s new o1 preview reasoning model.
Others have altered the prompt to receive brutal criticism and honesty.
And still others have completely other ideas for questions you could ask the chatbot, such as requesting it “roast” you in the style of a Comedy Central special.
Regardless of which style question you decide to ask ChatGPT, or any AI chatbot for that matter, regular users might find it interesting, amusing, and potentially revealing to learn what the chatbot says it knows about you — and more importantly, it may inspire you to think differently about yourself today.
Altogether, the interest in using AI models to find out more about ourselves and our own habits reveals how much potential they have, far beyond simply assisting with work or school assignments. Indeed, as the generative AI era approaches its 2nd year anniversary (since the November 2022 launch of ChatGPT), this question and the others like it show just how much AI has become embedded into the fabric of our lives and society, and how the more we use it, the more interesting new uses for it people find.
Source link