AI bias – a picture is worth a bunch of words

An interesting piece from NPR looking at attempts to have a LLM create the flip side of what it likely was trained on. This is currently a weak spot for many LLMs, but given how quickly the field is moving, it will be interesting to see how these issues get addressed (if at all) and whether fixing one area breaks another. At the end of the day, “good enough” usually is good enough, and bias often gets a pass. More’s the pity.