Rewind, Revise, Repeat: The Art of Working Around AI Gaffes

So what should you do when an AI model makes a big mistake? I recommend correcting it as matter-of-factly as possible and trying to briskly move on. If you haven’t yet built up a lot of context in the conversation, it might be worth starting over entirely. The best approach - only offered by some AI tools - is to go back to the point in the conversation right before the mistake was made and head it off by updating your previous message. In general, I think one underrated AI skill is developing a “theory of mind”3 for AI: a sense of how AI really works under the hood, and the kind of communication patterns that are a good and bad fit for it. Demanding that an AI apologise for its mistake is an example of poor theory of mind. There’s no sentient creature there to morally judge, it can’t feel bad, and it will make it more likely to make further mistakes.

Original post: gurupanguji.com

Beyond the Blur: How Tinting Liquid Glass Can Boost Legibility

When you layer the translucent effect with a color, you still get a glassy look but the text and iconography atop them are more legible. I recently switched to using an orange liquid glass tint on my lock screen and home screen. It looks particularly great when paired with the wallpaper I am rocking which features an orange flower. But the best part is that everything is even more defined. They also bring further personality to the system.

Original post: gurupanguji.com