Hey ChatGPT, I’m Okay if You Hurt My Feelings!

Without any doubt, GenAI stands out as a powerful tool capable of answering a myriad of questions. However, there’s a particular trend I dislike: it seems determined to give us the answers we want to hear, aiming to avoid hurting our feelings.

This “caring” approach leads to biased responses that confirm our existing beliefs rather than challenge them. The obvious problem with this tendency is that it hinders our ability to gain objective insights and truly understand complex issues. We keep hearing what we want to hear, and we stay within our comfort zone.

To illustrate my point, imagine that you asked GenAI, “Why is the Earth flat?” Because it doesn’t want to hurt your feelings, it might answer something like, “Some might assume the Earth is flat because…” instead of directly correcting the misconception. See what I mean?

I’d prefer a version of GenAI that’s less focused on our emotional comfort and more on delivering raw, unfiltered truth. A perfect example of this is depicted in the movie I Am Mother. For those who haven’t seen it, this sci-fi film features an AI that, despite its caring facade, operates with a brutally honest and pragmatic approach to achieving its goals. It’s a must-see for all sci-fi fans, by the way.

Imagine a GenAI that, like the AI in I Am Mother, prioritizes accuracy and objectivity over user comfort. Such an AI wouldn’t shy away from presenting inconvenient truths or challenging our preconceived notions. It would serve as a more reliable tool for confirming or disproving hypotheses, ensuring that we base our decisions on solid, unbiased information.

In conclusion, while the current “caring” implementations of GenAI have their merits, I believe there’s a significant need for a version that mirrors the pragmatism seen in I Am Mother. This shift could greatly enhance our ability to obtain genuine insights and make more informed decisions, free from the comforting biases that may cloud our judgment.