Thanks for the nightmares Mo

Is it possible for something to be both sarcastic, and genuine at the same time? That’s how the title of this blog feels to me.

Having just finished watching the Mo Gawdat podcast episode I am at a loss for words. I’ve yet to do a deep dive into the world of AI, but I wasn’t really worried or scared of what’s to come either. I can’t say the same about the latter anymore. I like what Mo said about how when as a society we go through a crisis of some kind we follow the same trend, and he used COVID as an example. He said first there’s ignorance, then arrogance, then debate, then blame, then agendas and our own benefit (my tribe vs your tribe). This idea that we wait until these problems become an immediate threat until we start to take them seriously is all too true. We look back years later and see the mistakes we made and wish he had seen it sooner (interstellar anyone?)

If you are anything like me you recognize that there’s an issue with AI, and yet you feel like “Well I’m just one person, there’s nothing I can really do about it”, so then we revert to our day to day lives with it. That makes the discussion of whether to use AI in schools really difficult. This technology is being developed at a rapid pace, and so I don’t think it’s fair to anyone involved in the learning process to try to teach these tools when even we don’t really understand them. How can we expect teachers to be able to educate students on safe use of AI when we haven’t had the training ourselves. In Mo’s example we are essentially “The kids”. We’re in awe of this innovative technology and see it as very cool and interesting, but we’re not thinking about the long-term applications of it, or how it’s going to change.

At the same time, it’s not fair to students to not talk about it in schools. This is the future whether we’re ready to accept that or not. Ignoring its presence is not properly preparing them for their future like we should be trying to do. Not giving them opportunities to utilize and discover AI’s benefits will put them at an extreme disadvantage when they become young adults. Mo discussed that the fear for a while has been “AI is going to steal my job” which could be true way down the line, but in the near future he says it will be “Someone who knows how to use AI is going to steal my job”. Therefore, the choice to exclude it or not from the classroom isn’t really a choice at all. We have to include it. Of course, this then cycles back around to the previous paragraph where we aren’t well equipped enough to teach them about these tools, but I’ll try to break free from the never-ending loop I almost caught myself in.

In Henrik Kniberg’s video he refers to AI as an “Einstein in your basement”, saying that AI is wise beyond any human we know, but that it also can still make mistakes. Those mistakes boil down to our own prompt engineering skills. I feel like that specifically is what we should focus on teaching our students. Of course we still need training on that ourselves.

Over the summer I wrote a blog post about how quickly AI has grabbed hold of our world and how many different tools are out there. I described it as a “vast sea of AI resources”. I also talked about privacy concerns and how we’re always under surveillance. I had ideas in my head of what these would look like as images, and so I recruited the help of AI to develop images. Both of them took hours to create something remotely similar to what I had in my mind, and even then, they still didn’t match my imagination exactly. It was attempt after attempt of tweaks to my prompts, both big and small, hoping that the next try would yield the results I wanted. I used two different tools to create these images. Both sites only allowed for a certain number of free prompts before you were locked out, and for both options I found my picture with less than five free prompts left to use. Now you might say that was convenient or lucky, but I wonder how much of that was just acceptance that these were probably the best I was going to get. I mean I kept prompting the tools afterwards for something better until I ran out of free attempts on both.

I guess the point I want to get across in this story is that I don’t know how to best use AI. I’m sure many of you are in the same boat. The only way to change that is to practice. We may feel intimidated to explore AI on our own, we may feel defeated if we don’t get it right away, but if we genuinely want to improve with it then we must take that first step and try it. Try it over and over again. The same can be said for our students. They may not know how to approach AI tools outside of school, they may not have a desire to learn more about it on their own, but with it guaranteed to be a part of their future I feel it’s our duty to give them those opportunities in the classroom to explore it.