What Should You Know About ChatGPT and the Art Room? (Ep. 358)

Welcome to Art Ed Radio, the podcast for art teachers. This show is produced by the Art of Education University, and I’m your host, Tim Bogatz.

I am excited about today’s episode. Amanda Hine is going to be joining me and we’ll be talking about artificial intelligence, specifically ChatGPT. Now ChatGPT is a web-based chatbot created by a company called Open AI. They’re the same company that made the AI powered image generator called Dolly that you may be familiar with. There’s a story about someone winning an art competition in Ohio a few months ago, using that tool to create his work. And so that made a lot of waves. And now ChatGPT and all of its uses are beginning to make a lot of waves now. So we wanted to try and cover that in today’s episode. Now, it seems like everyone has an opinion about chatbots or artificial intelligence or ChatGPT or all of the above really. Some people are really excited about the possibilities and the potential.

Other people are apprehensive. They’re worrying about the negative ramifications that could happen down the road and some that are already happening, and a lot of people haven’t heard enough to really form an opinion. Now, if that’s where you are, if you haven’t heard enough, let me offer just a quick explainer, at least what I can with my understanding where it is at the moment. So ChatGPT is one of many, many language models out there and one of many similar technologies, but it seems to be probably the most impressive or the most advanced model most people have seen to this point. And you can think of it like a really advanced Siri or really advanced Alexa, and it’s called a language machine. And it’s been trained by people to index and repeat words and phrases and sentences. It doesn’t have intelligence per se. It’s just kind of regurgitating that information for you. And what I’m saying, I mean it doesn’t have intelligence, is that it knows how words are used, but it doesn’t necessarily know what those words mean. But it has learned how to kind of copy writing styles and how to have conversations.

And it continues to learn and continues to improve, continues to get better. And so ChatGPT and most AI powered language machines, they have a really long way to go. They can answer questions and sometimes answer those questions impressively, but most of the time answers are very basic and they leave a lot to be desired. There are times that they can even be flat out wrong. They can give you false information, but all of these language machines are just continuing to learn, continuing to get better. And I think that’s why so many people in the tech world and so many people in the education world see a lot of potential uses in the future. And so what are some of those uses? For educators, I can see a lot. ChatGPT can come up with vocabulary lists and write quizzes and draft emails home to parents and give you discussion prompts. It can even write lesson plans. Now does it do all those things well? No, not always, but like I said, it continues to learn and improve.

But there’s also the flip side. There are things that frighten people. So what’s scary about it? What are the concerns? As with all artificial intelligence systems, there are issues of bias and discrimination. A little later we’ll talk about a specific problem that I found, but those issues are pretty well documented. They’re the obvious concerns about plagiarism and students having chatbots their work and do their research for them. And looking at the big picture, we can’t always foresee the consequences that come with new technology and of course that people are concerned about that as always. But just in thinking through those things, those are obviously not all of the possible benefits. And of course not all of the existing concerns. We could do another entire podcast on the possibilities and another podcast beyond that on the ethics and biases and the concerns inherent with artificial intelligence and its uses.

But for today’s episode, we’re going to talk about some of our thoughts, our experiences, our concerns with this tool as it exists right now, and a few examples of what we found when we started some conversations with ChatGPT. So let me bring on Amanda and we will get that conversation started.

Amanda Heyn, welcome back to Art Ed Radio. How are you?

Thanks so much for having me. I’m great. How are you?

Wonderful. Excited to talk about, I don’t know, artificial intelligence, robots taking over the world, whatever the case may be, however you want to categorize it. No, I was excited you wanted do this podcast with me, but I was also, I don’t know… It surprised me a little bit because you hate robots, but for some reason you were just fascinated with ChatGPT. And so I guess my first question is how did you find ChatGPT? And secondly, why does it interest you so much?

Those are great questions. So yes, I hate robots, specifically the ones that look like dogs and they… Have you seen those videos where they try to push them over and then the dog robot crosses its feet over and maintains its balance? That is so terrifying to me. Roomba is fine. Anything more intelligent than Roomba, I am not into. So yes. Okay, so ChatGPT should scare me. I don’t know why it fascinates me, but I had heard rumblings about it on social media, but I really started digging into it in one of my course discussions. So I’m in a master’s program right now about instructional design and learning technology, and the course is about digging into new technologies, evaluating new technologies. And so of course this week we had a discussion about ChatGPT, and it was really interesting to read through everyone’s discussions.

I’m in class with a lot of people who work in higher ed, a lot of people who work in K-12. There’s a lot of teachers in my class. And the discussion was just so interesting. I think particularly because there’s such strong reactions to it.

Right, right.

It’s like people are initially absolutely not. Teachers are like, this is cheating. And then someone will be like, well, but you could use it in this way. And people are like, then they run to the other side, they’re like, oh, this is really cool. And then, I don’t know, it’s changing so quickly and I genuinely am excited about the possibilities that ChatGPT could provide in education and elsewhere. It’s just really interesting to me. And when I finally started to play around with it, I found myself very surprised that I was not scared. I was delighted.

Oh, yes, yes. I love it. Yeah.

What about you?

I’m kind of the same way. Well, yeah, I’ve just seen it all over the internet basically. And I’m surprised at how fast it’s moving. My wife, I think people who listen know that she’s a seventh grade history teacher, but she had a teacher training this week on ChatGPT and how it can be used. And she came home with a ton of great ideas for it. And so it’s just fascinating. Like you said, some people still are just saying know how no way. And other people are embracing it very quickly. And I don’t know, I was skeptical at first. I think I’m still a little bit skeptical, but at the same time, I spent 40 minutes yesterday just playing around with different problems. How deep does this rabbit hole go or how can I stump it? And so just intellectually, it really piques my curiosity. But I wanted to ask you, I know we talked about recording this podcast. You put something on Instagram just asking about ChatGPT, people’s opinions. What did people have to say about it? What were their thoughts?

Yeah, so I think it’s interesting. I think a lot of art teachers are, like you said, familiar with Dolly because that is the AI that generates pictures. But I don’t know that as many art teachers are aware of ChatGPT, but I think that is quickly changing. So we had wide ranging, excuse me, responses. So some people we had saying, I literally never heard of this. And most of their sentiments were like, I haven’t heard of it, I’m scared, or I’ve never heard of it. And now that I have, I don’t like it. Which is interesting because that was my initial reaction as well.

Yeah.

We have a couple people who have heard of it and really like it. And the one I thought that was most insightful is someone said to us, it’s only as smart as the user, which is something I definitely want to talk about during this podcast. And then we also had really insightful comment about how students are using it to try to pull one over their teachers and then how teachers are trying to use other tools to detect that. And kind of reminds me of hackers, right? It’s like you always have to be one side, one step ahead of hackers. It’s like it’s becoming this weird race to understand the technology and how it’s used. And so that’s a really interesting discussion as well.

Yeah. So can we talk about that for just a little bit?

Yeah.

Because I was just thinking about the consequences, how it’s going to… How it could change how some teachers are doing things, how they’re presenting information to their students, how they’re getting information back from their students, how does it affect how you do assessments, things like that. But yeah, I know a ton of teachers are worried about plagiarism, cheating, just all those other consequences. What are you thinking? What are you feeling about where this is going?

Yeah, so like I said, a lot of people upon hearing this is a chatbot that can write an essay about impressionism for you in three minutes, rightfully so are concerned about that. It obviously brings up issues of originality, potentially plagiarism, although can you plagiarize a robot? I mean, I guess. Cheating. And I think however that said, it does kind of sound like a robot wrote it right now. I think that probably will-

Oh yeah, absolutely.

We’ll get into some examples, but I think that probably will change quite honestly. It’s only going to get better and better. The person who wrote to us on Instagram mentioned a couple of tools, but one thing she said was one really easy way to detect if a student has used it for an assignment is to check the Google Doc history. Because what will happen is the kid will put it in ChatGPT, and then just copy and paste the whole thing in Google Docs. And so what you have in the Google Doc history is just like this essay appearing, there’s no editing.

Instantly shows up.

Yeah, it instantly shows up. There’s no version history. So I thought that was a pretty insightful hack plus one for the teachers there for thinking about that. But I don’t know, what have you been hearing about it?

Well, it’s just more what I’ve been seeing when I ask it about artists or ask it to write about impressionism, for example. It’s just the most basic elementary stuff. And yeah, like you said, it sounds like it was written by a robot and I think some of your smarter students might do that and then like you said, revise it to make it sound like their own or to change it. But there’s still different tools out there. I saw a lot of teachers talking about this tool called Chat Zero where you just… It’s similar to Turn It In or something like that where you can run it through and it can tell you the odds that this was generated by AI or things like that.

Right. But now there’s an AI where the kids can put it through ChatGPT, and then they can put it through another AI program to make it not sound like ChatGPT, so it doesn’t show up on Chat Zero. And all of these things have happened in the span of a week. It’s just… It’s wild how quickly things are changing.

Yeah, I don’t know. I mean, I would hope that I know my students well enough and I know their capabilities well enough to know if they actually wrote something or not. But not everybody works with kids all that often. Or if you have 400 students, you don’t know these things. But I mean, you can always go back to just pencil and paper for everything.

True. So now kids don’t know how to write, they don’t even learn cursive. It would take them four hours to write an essay by hand.

What was I going to say? But you were telling me one of your classmates asked something about what it could do to frighten you.

Yes.

Tell me about that.

Yeah, so in the discussion, I thought this was a really insightful question. She said, “Is there anything that ChatGPT could do that would really scare you to your core?” And I was like, oh, yes. Okay. So if it became a sentient being, the movies, I think obviously that would be terrifying. I don’t really think that’s going to happen. I don’t believe in that. However, what does scare me is the fact that tools like ChatGPT or other AI tools could potentially hinder creative thinking. I think we’re going to get into talking about how they can be incredible brainstorming tools, but I wonder if that ultimately helps or hurts people using them. And like you said, I hope that I know my kids’ writing style or what they’re capable of, but think about kids who are nine now. Are they going to be using these kinds of tools and how is that going to impact their innate skills? On the surface, it seems like, oh, it would be so bad. But then I think, oh, actually would it be so good? Because you actually do need some incredible critical thinking skills to use ChatGPT effectively.

So I don’t know, I think there’s so much that’s still to be determined. It’s a really interesting conversation.

Yeah.

Does anything really scare you about it?

No, I don’t think so. I don’t think… I mean it learns from the internet and on the whole the internet’s pretty dumb, so I think it’s going to take it a long time to really, really outsmart people. So no, I’m not terribly worried about that. Like you said, it’s a tool and any other tool, we need to know how to use it. We need to teach kids how to use it. And I think that’s where the opportunities kind of arise. And I know a lot of teachers have been afraid of like, oh, AI is going to replace all these jobs. It’s going to replace what we do and how we do things. And I mentioned my wife doing her PD day on AI, and one quote that she ran across there was that AI is not going to replace you, but a person using AI might replace you. And I think that’s a good way to think about it. I think just learning to use these tools is going to be important. Like I said, teaching our kids how to use them is going to be important as well.

Yeah, definitely.

All right. So I guess we both did a little homework for this episode. We played around a lot with ChatGPT as mentioned. And I guess before we get into the specifics, what did you think about ChatGPT or just AI chats in general? What are they doing really well and information that you’ve seen other people using them. What’s impressed you and what is limited? What are the things that they aren’t doing as well?

Yeah, that’s a good question. So I think that one thing we’ve mentioned is that they are really a cool tool to provide brainstorming support. So I have a kiddo who has a really hard time getting started on homework, especially open-ended assignments. And so I sat down with him and he is nine, and I was like, okay, this is only to be used… There are so many caveats. This is only to be used with a grownup right now. It’s only for brainstorming. We’re not doing X, Y, Z with it. But we sort of put his assignment requirements into ChatGPT, and then he provided a little bit of context and then we said, give us five ideas for this. Right?

Yeah.

And so it did really well at doing that. Now it did… It was something about an adventure story and it got kind of dark and weird.

Okay, yeah.

And so then we revised and we were and keep it child friendly and then it revised and it really did come back to something appropriate for a nine year old. So that was kind of interesting. So I think brainstorming support is awesome, giving ideas because then he took those ideas and put his totally own spin on it. His final story was much different than the start, but it gave him that sort of entry point into the assignment, which was really interesting. I think it’s also interesting to think about how it can support those with learning differences. So I saw this woman online and she has ADHD, and she was talking about how she was using ChatGPT. She was putting in the transcripts from her meetings and having ChatGPT spit out a list of action items that she had to do.

That’s really nice.

Isn’t that brilliant?

Yeah.

Oh my gosh. I just thought that was such a cool way to support someone with a very specific need. And it had me thinking about could you put in, for example, an article and have it spit it back at a different reading level or something, or simplify a concept. I think that’s really, really interesting. And then the final idea that I heard that I thought was really cool was almost acting like a tutor. If you are trying to do a really difficult statistics problem and you don’t have your teacher there, could you use ChatGPT to say, can you explain this problem to me and how you would solve it in-

Yeah, that’s interesting.

Accessible language or whatever. So I don’t know, do you have any other ideas or seen any other cool uses?

I’m very curious about that last one and how well that would actually work.

Yeah.

Again, just trial and error with a lot of this or it’s still learning. We’re still learning, but that intrigues me. Yeah, I think the capability that I appreciated the most was like you said, just sort of the differentiation part and just thinking about speech to text or text to speech or however you… All these adaptations that we want to make for our students. They can really help us. But I don’t know, just the overarching theme for me and what you’re thinking about is this needs to be looked at not as a final tool, but as a starting tool.

Yes.

Like you said, whether it be for research or for creative prompts or wherever you want to go with, this is a good place to start, but we need to understand our students need to understand what it brings back is not and probably will never be a final product.

Right. Yep. That’s really important to understand.

So anyway, that is one of the limitations that really bothered me. It’s very basic on a lot of levels, and we can talk more about that, but what are the limitations that you’re seeing, or where do you think AI is falling short in these chats right now?

Yeah. Well, I think one of the biggest problems is you don’t know if the response you’re getting is correct.

It’s true. That’s true.

You have to be the filter to understand enough to know whether or not you’re getting good solid information because it will give you an answer no matter what. And sometimes it does tell you that’s incorrect. I think I put a prompt in that said something like tell me about pop art created during medieval times just to see what would happen. And it said, it did give me an answer. It was like there was no pop art during medieval times. This is what pop art is, this is what medieval art is. And then it tried to bring the concepts together, which failed. So that was interesting. And again, like we’ve said, it doesn’t sound super natural I’m going to say yet, because we know tech capabilities do increase exponentially. So I think it’s going to get there, but it’s incredibly long-winded as well, just-

Yes.

It just gives so much information. And I don’t know if that’s because it’s searching the entire internet, which obviously has gobs of information, but that was something really interesting. It’s not succinct and it’s not super natural.

Yeah. There are quite a few answers that just sort of ramble on and on and just give you way more than… Not only more than you need, but way more than you ever wanted.

Yes, yes.

What am I going to do with all of this? Or I asked it to write me, what was it, a song about some… I don’t remember. And it was like six verses long and five choruses, and it was like, nobody’s listening to this 14 minute song that you-

You didn’t ask for a musical.

Right. So it can be a lot. Okay. But I think we should maybe share some of the prompts that we put in there and just kind of talk about the results. This is part two of our homework here. So you want to go first and share one of your prompts, or you want me to?

Yeah, sure. I can go first.

Okay.

So kind of echoing back to some of the things we thought it might be good at, such as providing modifications or accommodations, I actually gave it a prompt that might have come from a student’s IEP just to see what it would give back to me.

Okay. Yeah.

Because in the art room, we see everybody. And so it’s really important to be able to differentiate for all different types of learners and students and abilities. And so I asked ChatGPT how would I modify an acrylic painting lesson for a student with limited mobility in their hands? And honestly, it was pretty good.

Okay.

So again, it gave me a five paragraph essay, but the headings were great. It said provide alternative painting tools. It said adapt the painting technique to maybe something looser or less realistic. And it also told me that I could adapt the painting surface, putting it at a vertical or an angled surface. And then it said also to break the lesson into smaller steps if the student was getting fatigued. So I thought that was really good. And then the final piece of advice it gave me, which was interesting, that it was giving me advice in addition to concrete steps.

Right.

It said ultimately the key to modifying a painting lesson for a student with limited hand mobility is to be flexible and open to experimentation, which I honestly thought was-

Nice. That’s good.

able to work. Yeah. So that was a successful experiment in my mind. Okay. What did you ask it?

Well, it started simple, and then it went into that rabbit hole like I mentioned earlier. So I just started with drawing prompts. I was like, can you give me 10 drawing prompts about animals? And it came up with 10 basic, like we said, very elementary level, but I would probably use most of them in my classroom there. They’re not terrible. And then I was like, what about animals and food? And then it got a little creative, a little weird. Koalas were making tacos, but the tacos were filled with even smaller koalas.

I want to see that. I want a kid… I want to see a kid draw that. That’s amazing.

Yeah. So there was some weird combinations, but some fun ones too. And then again, this is the rabbit hole that I talked about. I’m like, oh, can you take drawing prompt number three and turn it into a lesson plan? And then it gives you a whole lesson plan using those prompts. And it’s all about just, most teachers would write a lesson plan. Here’s what you start with. Here’s your activity, here’s how you wrap it up. Just like I was impressed with it it knows how to write a lesson plan, how to structure it. And then they’re talking about selecting paintbrushes. And so I was like, oh, tell me more about these paintbrushes. And then it gave me a breakdown of six different paintbrushes, and this is what you use a fan brush for. Round brushes are best for this. I was like, cool. And then what was next?

Oh, I asked it. Oh, you mentioned scumbling. Can you tell me more about scumbling? And it gave me a good description for scumbling. And then I was like, oh, who are some artists who use scumbling? Gave me a list of five. And I was like, oh, tell me more about that artist. And they did. And again, I’m not sure how great all of that information is, but it’s there for you. And then I was like, oh, who are some similar artists that might also be interesting to my students? And they gave me a list there. And so it really just kept me going and going and going. And like I said, before, I knew a lot of time had passed, but again, it was all basic information, but it was pretty thorough. I was impressed at how you could just keep developing from that original prompt.

Yeah. Well, can I tell you about the lesson plan I asked it to write for me because mine was not so successful, but I think that was-

Okay.

Partially on me, right? Because we were talking about it’s only as smart as the user. So I also went down a rabbit hole, but I tried to see what it would do if I gave it something not great as a starting point. So I asked it to write me a ceramics lesson for high school aligned to Wisconsin State standards. And then I asked it to write me a collage lesson for kindergartners aligned to the National Core standards for the National Arts standards, I should say. And those were fine. I don’t know. I would give the lessons a C minus. Could you technically follow them? Yes. But for example, it told me to have kindergartners write a report on Matisse as an extension activity.

Okay.

And it told me to have the high schoolers complete the entire ceramics lesson in one class period. I mean, again, you need to know what you’re doing. So then I was like, okay, what is it going to give me if I give it something really weird? And I said, can you write me an art lesson plan for eight year olds that they could complete in 50 minutes on the topic of hamsters?

On hamsters?

And it gave me a lesson called hamster portraits. And in the introduction it told me to show students pictures of hamsters and talk briefly about their unique features and characteristics.

Oh, I was afraid I was going to have it bring it, have your students bring in their own hamsters.

Okay. Well, it did as an extension activity, it said, but I thought this was good. It said, invite students to bring in their own stuffed animal hamsters.

Okay, okay.

And take a class photo with their portraits and stuffed animals, which I thought was good for eight year olds. And it got into some interesting things about texture of the fur and stuff.

Okay.

But I don’t know that we’re going to be having ChatGPT write our lesson plans for us right now, in fact.

No, no.

I don’t think… I could not recommend that at this point.

No, it has a ways to go. I’m intrigued by the possibility, but yeah, it’s definitely not there yet. Just a quick aside, I did ask ChatGPT to come up with a vocabulary list for basic ceramics course, and it gave me 15 different items with decent definitions and all put into a nice little table for me. I was like, oh, this is good. So again, it’s a good starting point.

What’s really interesting, it’s if it’s crawling the whole internet, is it also crawling the art of education? Do you know what I mean?

Oh, that’s good. Giving me my own vocabulary list back.

Right. Did you write that six years ago? And it’s just giving it back to you almost correctly.

Was going to say, okay, so second thing I did, second experiment I did was I was just thinking what would be most helpful? What do I hate doing in the classroom? I hate contacting parents. And so I was like, ChatGPT, can you write me an email home to parents? And so the prompt I gave it was just, can you write me a parent email home talking about how their student is good at X, Y, and Z, but struggling with this topic and also needs to complete this assignment. And it just spit it out and it was ready to send.

Wow.

It even gave me… It put your name in big block letters to tell me where to put my name in and where to put the student name in. And it was ready to send. I was like, this is good.

Wow. That’s incredible.

That was a definite win. So yeah, I just went in there and said, the student has been great with behavior and working on these things and helping these students out. And it put it together very eloquently and then talked about these are things we’re concerned with and then ended with love having student name in class.

Student name.

Appreciate everything you’re doing to support them at home or something like that. I was like, oh, okay, this is ready to go. I like this.

Again, we have a PRO Pack on that topic and did it just take it from there? I don’t know. Interesting.

What about you? What was number three? What was the other thing that-

Yeah, the one other thing I did is I asked it to write me a rap about the Renaissance. And I don’t know, it was not that great. Mostly again, because it was a page long. But I would like to read you one of the verses.

I was hoping that you had maybe-

I do. It says, “Listen up y’all, let me tell you about the Renaissance. It was a time of art and culture, a period of great advance.” So again, back to doesn’t really know what it’s saying, Renaissance and advance do not rhyme. “From the 14th to 17th century, things were changing fast and the art that was created during this time would forever last.” So it’s not wrong. It’s just not going to engage any kids.

No.

It would be kind of funny. I could see an interesting lesson with high schoolers because they would find that cringe-worthy, but they would also find it very funny.

Right, right.

And if you were in an art history class and you had a prompt who can get ChatGPT to write the funniest rap about whatever I think that could be… I think kids would really enjoy it and they would be learning because again, the information spiting back is not, in this case incorrect, but it’s a pretty bad rap.

Yeah, no. Not great. You’re also going to have a class full of kids pronouncing it, the Renaissance.

Yeah, right.

That’s a stretch, but okay.

I know. What about you? What was the third thing you did?

Well, this is where we kind of ran into a problem just showing off some of its limitations. So we talked in the podcast two weeks ago about showing black artists not just during Black History Month, but throughout the year. And so that was on my mind. I was just like, hey, ask ChatGPT who are some famous black artists that students should know? And it came back with a list of artists, probably a similar list to what I came up with, but the bio for every single one, for all five of them used some form of this artist’s work deals with race, identity and representation. Or the next one would be like, this artist’s work deals with representation, race, and identity.

Oh no.

And so my follow up question was, do black artists only deal with race, identity and representation? And then it just spit back a word salad of, I don’t know.

Yeah, not correct.

It had no explanation or definition. And I think four of the five artists were also… Had a tagline of they had won numerous awards and been recognized for their work.

Oh my God.

The most general basic stuff. So anyway, you can submit your feedback. And I was just like, you’re struggling with this right now. And I think that goes back to kind of the implicit bias that I talked about in the intro to this episode. And these tools are only as good as the people who created them. And it’s something that we need to be aware of and something that we need to probably talk to our students about. But it’s definitely, yeah, like I said, it’s limited in a lot of ways, but I’d be curious to know how fast it can learn, how fast it can improve and do something better.

Yeah. Well, I think that’s one other important point to make with this, is if you use ChatGPT, there are people behind it and they are studying what you put in and what it spits out. And it does give you that disclaimer at the beginning, not to put any personal information that you don’t want ChatGPT to know in ChatGPT. So yeah, again, it’s just everything’s changing really fast. And you have to be put some guardrails and some parameters around how you use it.

Yeah, yeah, for sure. Okay, so Amanda, to close the show here, I thought it would be fun if we give ChatGPT a live prompt and just see what pops out of there. So I have ChatGPT up here. What would you like me to do? What would you like me to request?

Okay. So I think the funniest things are a poem or a song of some kind.

Okay.

So do you have a topic in mind?

Well, I mean something art teacher related. Can we do an appreciation poem for our teachers?

Yes.

Okay. All right. So I’m going to type in can you write an appreciation poem for art teachers? All right. Okay. ChatGPT says, “Sure, here’s a poem to appreciate art teachers.” Just any predictions on what the worst rhyme is going [inaudible 00:36:01]?

Maybe art teachers and bleachers or, I don’t know, but I’m assuming it’s going to be lengthy and thorough.

Yeah. And I think some lines are like, wow, that’s really good. And then other lines are going to be absolutely cringe worthy. But here we go. “Oh art teacher. How you inspire with color, shapes and lines you conspire to show us how to see the world anew and create beauty that’s fresh and true.” There are five… Are these stanzas?

Yeah.

I’m not up on my poetry. I’m sorry.

I think they’re stanzas.

All right.

Yeah, we’re not English teachers. I don’t know.

“You guide us through the creative process from sketching and shading to choosing colors with finesse.” Oh, there we go. Process appreciation.

Yeah, no.

“You teach us to trust our instincts and take risks and help us find our own unique artistry mix.”

Okay.

“Through your patient guidance and skillful teaching, we discover the joy of self-expression reaching.” That makes no sense to me.

No.

“We learn to look at the world with a different eye and find new ways to communicate, to express, to fly.” Getting vibrant with the language I like that.

Yeah.

Okay. “With every stroke of the brush, every smudge of the pencil, we feel the magic of art, a creative vessel.”

I do like the magic of art.

“And we know that without your support and dedication, we wouldn’t be where we are with newfound imagination. So here’s to you art teacher with gratitude and praise for opening our minds and hearts in countless ways, for sharing your passion and inspiring us to be the artist we were always meant to see.”

Okay. Well, that is nice.

That’s a good ending.

But it gives me good feelings, honestly, about what I expected.

Yeah.

I do like how they made it from the point of view of the students.

Yeah.

That was a creative choice.

Nice.

By ChatGPT.

I like it. It’s fun. Yeah. There’s a couple, like I said, cringe-worthy pieces, but a couple redeeming qualities. Not bad. Should we try a haiku really quickly before we go?

Yes, please. Could I try one?

Yes, please do.

Okay. All right. What should it be about? Can you write a haiku?

I would say this.

That’s the haiku. Okay.

Art teacher haiku.

Just art teachers?

Yeah.

Okay. Are you ready?

Yes.

“Inspiring brush strokes, guiding hands with gentle care, art, teachers, masterpiece.” Not as good.

Completely nonsensical.

I mean. Yeah. No.

I’m going to say that’s a fail. That’s a three out of 10. Yeah.

But I mean, you can see how this absolutely becomes a rabbit hole of things you can do both for knowledge and entertainment. There’s just so much.

Totally. And I would encourage you, if you’re feeling apprehensive about it, try it out. Because I was, again, from my personal biases towards robots, very skeptical about ChatGPT and I have really come around and it’s honestly fun to play with. So I would encourage you to check it out.

Yeah, no, that’s excellent advice. But Amanda, we should probably wrap it up before we dive too deep into art teacher poetry. But thank you for this conversation. I know it’s been a long one. I know there’s a lot to talk about, but I appreciate you coming on.

Yeah, thanks for having me.

Okay. That was a very long episode, but we covered a lot of ground, and even though I feel like we talked a lot, there’s so much more out there. And as I said at the beginning, we could do multiple additional episodes on the same topic. But for right now, I would love to hear what you are thinking about ChatGPT, about AI, about any of these chatbots and what the possibilities are or what the consequences are in your own classroom. So if you have time to email me, Timothy Bogatz@theartof education.edu, I would love to hear your thoughts, your concerns, your opinions on where any of this is going. Are you using it? Are you scared of it? Do you want to learn more before you dive in? And if you’re already using it, I would love to hear what you’re using it for.

And Amanda and I were brainstorming some possibilities, but I feel like there’s so much more out there. So I’m very curious to know what people’s opinions are and if they’re using it, how they’re using it. So yeah, if you have time to reach out, I would love to hear any of that. Otherwise, we’re going to wrap it up for the day. I hope we kept you interested through this pretty long episode. I’m fascinated by all of this technology and everything that it can do, but like a lot of other people still concerned as well. So yeah, share your thoughts with me if you have the time. But until then, we appreciate you listening. Art Ed Radio is produced by the Art of Education University with audio engineering from Michael Crocker. Thank you. Good luck with ChatGPT, and we’ll talk to you again next week.