MAX SESSIONS Generative artificial intelligence has been a hot topic of conversation in classrooms and schools all over the US and all over the world. And it has teachers and students and just about everybody wondering what's its place, where is it going in education and how can we make the most of it? And so in today's session, John and I are going to talk a little bit about that, about that idea of generative artificial intelligence and its potential in education. And, the genie's out of the bottle, so to speak. Pandora's box has been opened, and so it's time for us to take some steps and to start trying to reimagine what education is going to look like with generative AI as a big part of it. John, I've been thinking a lot about it. I know you've been thinking a lot about it. Where's your mind when it comes to the potential of Gen AI in education? Oh, my gosh. So you say, "We have 30 hours for this session, right? -Not 30 minutes, right? -Right, exactly. Thirty minutes. All right, we'll do our best.
Yeah, it's incredibly exciting. So a little background for me that may shed a little bit of light. I started getting into, kind of, the space of what can educators do with tech, back in the 90s. You and I were talking about college in Indiana and, making it up as we went. I had the good fortune of being a freshman when really the Internet as we know it, kicked off with Mosaic and Netscape and I remember calling my parents. I'm like, "We talked to a computer in Czechoslovakia," and... When there was a Czechoslovakia. Anyway, I got into trying to work with professors to embrace whatever promised new changes, being able to stream audio. The fact that you could even record a lecture seemed crazy or put it online and there were a lot of the same anxieties then, differently framed like, "Well, if I do that, why would people pay for school or why would they even come to my class? Won't they listen later?" And so, I think any new technology always brings those disruptions. Obviously, Gen AI is a whole new ballgame, in terms of being able to synthesize things and so quickly.
But I also think that what I'm interested in hearing your thoughts on and talking with folks about is how do we negotiate that tension where a lot of what has been, sort of, high-touch craft maybe doesn't have to be, at least in the same ways, but the journey is still the reward, right? And a lot of value still is in the struggle and is in doing things which are meaningful as a human, because they reflect, sort of, hard-won insights. And I'll get to this in a couple of minutes, but what makes you human? And based on that, what do you bring to the table that a computer doesn't and how do we work with those things together? So anyway, like I said, "I could go 30 hours." I'll try not to, but yeah, this should be a great conversation. Yeah. Yeah. I've already heard versions of that from the teachers that I've worked with, whenever I've gone to, workshops and worked in schools and school districts and everything. And they're concerned about students still being able to think and reason and solve problems and bring value to the world. And, when a student decides to take an academic task and try to so-called, "Farm it out to AI." -Right. -Run it through one of the AI assistants, instead of actually doing the thought themselves, it runs up against that concern of the future of... If this is how you're going to treat your education, then what happens when you get out into the workforce? If that's all the thinking that, that you put into this, like this is the kind of thing that's gonna get outsourced pretty quickly. And so how do we maintain that as humans? And that, sort of, rolls us into our first question, "What skills are uniquely human or best done by humans, that we can encourage students to pursue?" It's almost, like, with all of this artificial intelligence here, and how it's going to continue to evolve and become more and more widespread, it's almost like trying to future proof or protect ourselves as we go forward to ask this question. And it's something I've been thinking about a lot. John, what's your take on it? I think you touched on something that seems important to me, which is problem solving and creating value.
To be a graphic designer before the 80s, for you to solve a problem...
some of the things were the same, some weren't. Right? Obviously, understanding a client's need or what you're trying to communicate, those things are, evergreen. But a lot of the mechanical parts that were part of the problem solving have gone away. I have a son who's turned 14. He's starting high school. He can't wait to take graphic design. And he found a drawer of old letters that you would have to cut out and rub down. And it's funny to talk with him about this notion that, that was the font, and you had to be super physically careful cutting it out and putting it down because you only had so many capital letter C's. If you run out of those, like, "Oh, Where can I get some more?" And so, we would think that's insane to go back to that. So the fact of some of the mechanical pieces going away is okay, because it does... it then did free people to go up the value chain. Right? Like, "Well, I'm still... I need to compose something beautiful and that it has clarity of message." The computer at that time certainly wasn't going to do it for you. So you still bring that in as a human? I think... I don't know, a lot for me, sort of comes down to be interesting, which sounds, I don't know, kind of glib, but it's like, "How did you and I get here?" I think of that famous story about Picasso drawing something, a lady came up to him and said, "Oh, will you draw me a bowl or whatever?" And he grabs a piece of paper and it turns around, but that'll be $10,000. And of course, he's super offended because of, "How dare you?" And, "It only took you a moment." He says, "No, ma'am, it took me my whole life to be able to draw that." And I think it's a nice shorthand for how you spend your time. your friends, your loved ones, all the things that got you here. This is part of why having people with diverse backgrounds, however one defines diversity, is so interesting and powerful, because just the road that got you here is what informed your taste and your sense of humor, your sense of style. I think those are the things where, as we're seeing, computers can now approximate that. you can say, give me a poster in the style of Shepard Fairey and it knows who Shepard Fairey is and then he's like, but also meets Norman Rockwell and maybe Kafka and you're like... So it can do those things, but you're still the one in a position to care, right? Or to have the interest and also, frankly, the judgment to know whether this actually solves a problem. I mean, I think people are desensitized so quickly. I know I get there, just seeing like this, impossible flood of beautiful things on Twitter and Behance and anywhere you'd care to look. Now, are they meaningful? That is kind of the thing. It's like, this is a longer, again, existential, sort of, inquiry, but, what is it that humans define as meaningful? And then how do we educate students to seek that, right? So as you're actually solving a problem, you're actually addressing a need. The need is not just to make something beautiful, because that's trivial, anyone can do it now in moments, but rather it's like, I'm currently trying to design artwork for the new milk truck that my brother bought to start a milk-hauling business. And it's still the dialog with him about what's meaningful. It's not cows, it's, why is it three cows? Why is one lying down? Why does it have a green spot? Oh, these represent my kids. And well, that's something meaningful about this, the one kid who's different from the others. Okay, cool, the mechanical execution is fun and it's... and to be able to explore at a super human rate is amazing. But it's still my job, as a person, somebody who knows this guy, to have that dialog. Yeah. Yeah. No, I'm with you on that. I thought it was interesting. So part of what I'm hearing you say is that it's just sort of like our human drive to create and to make things and to bring things into this world and to have value and everything. And that's why when I considered this question of what skills are uniquely human, creativity was one of the first things that popped to mind. Although I do know that, there are all these different tests of creativity and people are starting to show that some of the generative AI can meet to some of that, but I still think there's something about human creativity. For instance, you could go to Pottery Barn and you could probably find, like a some sort of piece of pottery that has been mass produced, and you can probably get it at great value. Or you could go have an artist create something for you, that's very different than what everybody else has. And there's still a draw to have that and there's still a draw for people to create it. You know part of it is, I have something unique. Part of it is I know the story behind it. -Right. -And so, those are some of the things that I think we best do as humans. It makes me think of there's the Japanese term wabi-sabi, which has to do with imperfection. I heard somebody explain it how, an artist using wabi-sabi would create something, a painting or sculpture or whatever and then would introduce an imperfection to it on purpose. Sort of like show that, that is the, sort of, the human nature. That's what shows that it's real. I've actually found myself in emails, sometimes not going back and fixing my typos because I feel, that's a way that people realize that there was a real human that actually composed that. -Yeah. -And so, I wonder if that humanity is part of what we're talking about. I think it is. That brings up a bunch of things, sort of ancient and modern. I remember in even Western traditions, artists would put a flaw in the same way, because they didn't want to offend the Divine by trying to say, "We're perfect. We can even get there." So they would deliberately take a step back and put a stitch out of place or put an insect in a painting to say, "Hey, memento mori," right? This is beautiful, -but this fate awaits all roses. -Yeah. I also remember when Snapchat blew up, it felt at the time, like let's say, 2010, 2012, thereabouts. Like, "Well, haven't these problems already been solved?" To the extent I have a problem, which is I want to share photos with the world, Like Flickr did it, and then Instagram made it even simpler. Those things got so good at letting you present this hyper-curated vision of life that the pendulum swung back and people said, "No, I want the real. Rough is real." And to your point about whether it's a typo, whether it's the fact they didn't put on makeup or shave or whatever it might be, people treat those things as more authentic. Now, you even see brands, you know... I saw Gwyneth Paltrow yesterday recording a video with horrible audio and her son is steaming milk in the background. And on some of them, I'm like, "Come on!" Of all people, you are the avatar of conventional perfection, but like trying to put out this rough aesthetic because it's so accessible to put on the perfect face, right, that people, they almost respond like it's empty calories. And I see this with AI-generated art. It's like, yes, you can have this sumptuous background that you would have had to commission a human artist if you could ever find one.
So I think naturally, people... It's like, the Grunge movement, right? These things, people swing back because there is that desire for the authenticity which comes from lived experience. Yeah. Yeah. No, no, I think you're right on there. And, I think since we're framing this in terms of education, I still see the relationship side of things, the student creation, the not... the knowing that, "Oh, this wasn't spun up by a bot. This was actually made by somebody that I know." It all sort of ties together. So a lot of that stuff that we do in the face-to-face classroom, even though, we have AI that can do some of that, I think is, it's still something that's relevant to what people want and what's best for us as humans. And, all of this, sort of, rolls us into this next question, which touches on similar things, which is our role as creators, which again, is a big part of what education is and that is, "How does the role of the human adjust, (and to what extent)... does it adjust, "...from creator from scratch to editor?" Do we make everything from scratch? It's just like what we were talking about earlier, the pot or the sculpture or the art or the classroom creation or whatever. How far should we go in, in that direction? Yeah. So it's interesting. My other son is interested in design, but in different ways than the younger one. And, apparently, there's a thrifting club at high school. I had no idea this was a thing. And I'm like, "Oh, like saving money." He's like, "No, I'm, going to thrift stores and buying clothing." "Okay. That's cool, if you find fellow travelers." But I mention it because it's a manifestation of creativity and of defining one's identity and exploring that with others. But, if you think about the mechanics of it, it's like being driven to a store and picking out items and putting them on. And you think compared to, like, most of human history, where, having the color purple meant you were royalty, because it could only come from this particular shell, which was hard to get and had to be crushed up and taken on some spice route. It's like, "God, if you can have a purple T-shirt, you must be the king," and they'll dig up a burial site and it's like, "Gosh, this little girl was buried with these, like, intricate whalebone trinkets and she must have been super high status." Whereas now you think you go to the thrift store, you get the whalebone or some approximation of it trivially cheap. But people redefine. They move the Overton window or however you want to call it, and now it's not, "Well, is there one bracelet?" It's like, "How did I editorially engage with the media at my disposal?" And so whether it's like high-school kids putting on thrift store stuff and then mixing it with some expensive sneaker.... Too expensive, by the way. But don't let them get started on that. I think they're all humans, adapting and re-leveling. And what I suspect with AI is things that have been really hard become accessible, but it's okay. People change their expectations. I mean, I worked on Photoshop for ten years and I remember when not... I mean, not only Layer Effects came in, when Layers came in. I think you're old enough to remember, but prior to that, Drop Shadows were cool, because you had to be such an alchemist and it was a tightrope act to like, "Make the selection and don't deselect, because you got one Undo and if you do that, oh, my gosh, I'm screwed and I have to start from scratch." And so, the fact that you could go through this gauntlet to get something with a piece of text with some Drop Shadow. There's a reason Alanis Morissette's early '90s album art has that thing because only high-end artists could make such a thing. Then it became trivial and every joker would put it in their church newsletter flyer. And so, some of the aesthetics we had valued became cheapened, but it then makes people step back and say, "What do I care about, though?" It's like, "Sure, is it like, this fantasy art dreamscape?" Or, "Is that relevant? Or maybe I should step back and think about what we're trying to say here." Yeah. Yeah, I. Yeah. No, I think it's interesting how... I think you put it so nicely that it's almost saying hard things become easy. And so then, the question then becomes, if this was hard, but now it's easy, what's hard? -What's worth doing? -Right. It's that whole perspective. And I think in the classroom, again framing this through the lens of education, there's a big... I can't tell you how many times I bring a statement like this up, going from creator-from-scratch to editor and you can see in the eyes of the English teachers. There's fire in them a little bit because all they've known their students to do is to be creators from scratch or a version of creators from scratch. And now, if we're going a little bit more towards the editor side, it's tough to reimagine it. In fact, this week, I was in a workshop and one of the English teachers piped up and finally said, "I want to know what my kids think. I want to know what they say. I don't want to know from an artificial intelligence." To which I started to think, "Well, we're already sorted down that path already," because before, think all the way back to the printing press. If the version of, "I want to know what my students think?" would have been, "What can they say out of their heads?" Now, we're way past the printing press. Now we have books and now we have the Internet, now we have... -Right. -And what is modern research? Modern research is taking lots of studies and pulling them together and making a point out of it. -Right. -That's not creating from scratch. -That's editing. It's like... -Right. ...we've continued to have these evolutions in human communication to this point and now it's going to be interesting to see what human communication, through the lens, through the scope of classwork, is going to look like. And so, does that mean we have an AI-generated essay and students improve upon it and then talk about how much they've improved upon it? -Yeah. -Do we've the AI spin up some ideas, then I create something from there. Like that's where we are right now with classwork, I think. Yeah, I think you're right. So I'll give you one example that I found kind of magical in its own way. The fact that you and I are talking to each other from multiple time zones and recording and sharing with people around the world is... I don't get... no retroactive stage fright saying this, but that's a miracle, right? But we take it as, "Oh, this is called Friday and it's whatever PM." But I had a conversation with a guy named Sanjay who's a PM on Illustrator. This was even several months ago. And I thought of this because we were mentioning, using AI as an ingredient. I'm overwhelmingly thinking of the visual components, but we were talking about logo design and what would it mean to have ChatGPT or Bard or something similar functioning as a co-creator, where it's not making the thing for you, but it is... it's something with which you can have a dialog and it'll take you in directions you wouldn't have thought of, because it is doing something human-ish in that, like you say, it's taking a lot of data points and then synthesizing something from them, which again is... that's why we liken it to human intelligence. But he and I and the fact he was in India. And again, this was part of the miraculous-ness to me. I'm sitting here on my couch at 11 P.M. He's there at 10 A.M., whatever it is, we're talking to a robot, wherever it might be, ChatGPT. We're asking it to come up with ideas for a health clinic for doing cataract surgery. And then we're thinking about, "Well, okay, if I now paste that into Firefly, what happens?" And, "What would that look like as a tool experience?" It was very memorable and exciting, because I think it points to a direction where you the human aren't marginalized, you're not edited out of a job. You have these new modalities to help you try more things. Maybe you literally do a napkin sketch. You can take a photo with your phone. We can pull it in and say, "Hey, do you want us to straighten the rough lines?" "Sure. Do you want us to vectorize that? Okay." Maybe then you take it and colorize it and you're like, "Hey, I did a sketch of the golf club at this resort." They want 20 other things. They want to see what would this look, if you did it for the pool? And what if you did it for the pro shop? And it's like, "I could in theory, with world enough time, sit there and do 20 copies of each one times five different styles." I'm not going to. Mechanically, I cannot, it's not economical, I don't have the bandwidth. But what if you as the editorial voice, could take that seed that you created, maybe with AI's help, maybe not, doesn't matter and then visually extrapolate out these variations and come up with an entire design system and get quick feedback on that, in a way that better solves the problem than you could have given the constraints of previous tools. Yeah. Now, what's interesting to me about that conversation you were having of finding ways to save time and be more efficient, sometimes when you bring this down into a classroom, you have teachers that will look at this and go, "But how far is too far?" Which leads us to our next question, because academic integrity has become a big discussion, now that more and more AI is coming into the world and is starting to find its way into the classrooms. And you were talking earlier about what would it look like if AI is a thought partner. And so that's, that's assistance. But if you take it so far as that it's starting to do most of the work for you, that's the concern that I know a lot of educators have, is getting that balance right. And, of course, the more that AI starts to become widespread, the definitions of some of these words, like cheating and academic integrity, are going to have to morph. It's like what happened when search engines started coming around, the teachers were all worried, "What if the kids look up the answers?" Now, we've evolved our classwork to the point where it's not about regurgitating answers, it's about doing something with those answers. And so that's the next evolution that came in education. So now, the question becomes, how are our definitions of cheating and academic integrity gonna change, -now that we have all of this AI? -Yeah. What do you think about all that? I think it's a very fluid situation. This week I was chatting, I was calling them the Kevins, the Kevin Barry I worked for at Notre Dame and Kevin McMahon, who's a design teacher at my kid's school, and having these interesting dialogs. I don't think any of us purport to have it all sorted out, but these were some of the questions we were raising. Is there a difference between using AI in the sense of Generative Fill in Photoshop, where, in the base case, you're removing something. It's like, that's content [inaudible] not, it's doing a better job, but it's not holistically changing things. And like you say, all the way down at the other end, where you're having it write a script and generate storyboards and not many months from now, probably synthesizing video and editing it together. I mean, I still... I think a couple of things. One is, rules of the road, right? Know clearly with for yourself and for students, like...
"How do you differentiate uses?" Two is, the goalpost will move, right? Things that are today's equivalent of cutting out those letters and rubbing them down and... Gosh, I couldn't try a different aesthetic because I don't have the letters, which is insane. That goes away. Great. Now you can try five things, but are you still building the muscle...
that helps you as a human, like, you know, work with a client, let's say, right? Like ask the questions and then question the answers. Right? Like teach the five whys of, "Well, okay, you want a brochure?" "Why?" "Because, I need to sell it." "Why?" Right? Like, "Can you develop a practice where the student opens more daylight between, what they as a human understand, and how they can empathize and systems which are better simply at, correlating a lot of points and producing an artifact." And that way, I hope as long as it's clear to everybody, like, these uses are okay and this is a bridge too far and you're treating it as an ingredient...
I hope it ends up in a good place. But I can't say that in my conversations, people have drawn bright lines. I still think... I mean, even, Fall semester versus Spring, we might see changes from doing an exam in class or a written thing at home, to an oral exam. Right? Because, ultimately, you're testing... Can this person think on their feet and show mastery and solve problems? Right? And that's... Then, whatever the tool is today or five years from now or ten years from now, those are the durable skills, right? Having the live perspective and the rigor to operationalize it. Yeah. And I think what you said, "There was key." Are we able to show skills? Are we able to show learning, all right? And all of that has been fluid and I think like you said the goalposts will move, things continue to change, and I think one of the keys that we have to continue to remind ourselves is, "Are we preparing our students for their future instead of today?" Because today was our future... -Right. -...but today is their past. The AI that they're going to work with today is the most rudimentary that they will ever see in their entire lives? -Yeah. -So, I think we've got to continue to think into the future and go, "What is this going to look like for them? What's going to be, our earlier question about creating from scratch versus editing. Because we create from scratch to do a certain thing today doesn't mean that we won't have AI spin up a draft and we do something with it in the future. And so little things like that, I think, are key. And it's something you said earlier made me think. We're always riffing, right? I mean, I think it was a YouTube channel, Everything is a Remix and Lawrence Lessig and people who are advocates of more permissive copyright interpretations, make that point. Rap culture is, because people took funk and disco -and found the breaks... -Yeah. ...and mixed them and then people came up with breakdancing, because that's like, oh, something you would want to do in response to this new distillation of an existing thing. And I think we want to find the new breakdancing. I think just stuff which was prohibitive becomes trivial, but that's okay, because there's still the same kind of distribution, I think, in people's creative ambitions. And those of us who want to lean into it. I mean, I was hanging off a cliff in Ireland a couple of days ago with a 360 camera on a ten-foot stick flying it around and then today, I'm like, "I wonder if I could, throw that into a 3D app and do a volumetric calculation." But weirdos, I say in the most affectionate way possible, are still going to be weirdos, right? Most people are to photography like I am to wine, which is there's bad wine and there's wine, right? And as long as you take a semi-competent photo, they're good. I learned this to my chagrin at Google, because most people don't want to edit. But there are still the students who are drawn to this world who... for whom I hope the journey is still the reward. It's not the outcome, right? It's you're doing it because the act of doing it, the act of engaging with the art and the problems is interesting. And then, as long as that North Star is true, then all the AI is doing is opening new thresholds, right? It'll redefine. Oh, for me, making a GIF banner ad in the 90s was something I could bill the client for all day and it was. I still have the GIFs and I'm so proud of them still. But obviously, that became trivial, as it should have, right? And then I moved on to, like, "No, actually design the whole website." That day, you got back. I don't know exactly how that'll manifest with AI, but I think a lot of the drudgery, to your earlier point, will be replaced by new creative thresholds that open up. Yeah. Yeah. I think so too. I think that's the optimistic vision that a lot of us have. And when we work and strive toward that vision and we sort of free up some of that bandwidth for other things, the question then becomes, what is it to be human? What's our best use of our time and our effort? And these are all of the fun conversations that we get to have when it comes to generative AI and its place and especially in education, but of course, we know education is preparation for the world. -Right, right. -So, with all of it, it's a fascinating road that we're all going to be taking. I think the best way for us to come up with the answers is to take imperfect first steps and try things and then share them widely, -which is what... -Yeah. ...especially in education I'm seeing people do, -which I think is super cool. -Yeah. And keep talking, right? Nobody at Adobe has any special purchase on the truth, obviously. Given we're all fellow creators, a lot of us came here because we were artists, photographers, educators, folks who care deeply about that space, and the only way we get somewhere interesting is through that dialog. So I'm jnack.com, jnack on all the socials, hit me up. I'd love to hear what people are contending with, and as with a lot of my teammates, and I'll help you connect with them. Yeah, yeah, absolutely. I'm open to the social media connections to @jmattmiller on Twitter, -which I guess is now called X... -Yeah. -...unfortunately. -Yeah. I am gonna be... That's a whole different session that we could do, isn't it? I'm gonna be the old guy who keeps talking about the Pan Am building and the Sears Tower, and it's like, call people by their maiden names. -Yeah, yeah, exactly. -There's no X in this vocab, so. Well, this has been a fascinating conversation, John. Thank you so much for your part of it. I've enjoyed it an awful lot, and for John Nack, I'm Matt Miller. We've enjoyed being here with you for this presentation. -Take care and we wish you the best. -Right. Thanks, everybody.