I wished they also measured/published the best estimate of resource consumption of these models.

Has it pedagogical value? Maybe. Was it fun? Definitely.

Something we work on from cradle to grave 💪

Indeed. I ran an (internal) AI training session recently - prompting techniques. "But what if students cheat with that?" - no no no! Ask why (some) students may feel the "need" to cheat! What may be the root causes for a specific assignment (in case there are problems)? Ask that.

I call it "the witchhunt mindset" 🧙

You could claude it 😂

Humanities have such an important - crucial - role in ethical AI development and use. Think what philosophy and psychology can contribute here! Cutting them is negligence!

YES! I keep saying in the AI trainings I run - "talking" to AI tools has improved my precision and clarity in how I communicate today. Less "reading between the lines", more face value.

A6 - Life literacy. In 5-10 years, GenAI will be as "normal" as smart devices are today. And we're still struggling to teach our kids to be literate with those. The most important skill: Learn to never want to stop learning! #LTHEChat

"Ipse se nihil scire id unum sciat" comes to mind 😊

I sign that, too. More often than not, we don't need to know 100% to move on successfully - 60%, 70%, 80% (I'm making up percentages) might be sufficient.

Most definitely. Personally, I've never been as creative in creating images/videos using AI tools than before (now I have a tool that helps me express my "artistic" side). The so-called "higher level skills" come to play and they're here to stay!