AI and Creativity: A Conversation with Tom Barrett

Last week I had a great conversation with Tom Barrett about the future of creative AI tools in education, including AI writers and AI image generation tools. We’re both optimistic about the potential for these technologies in the classroom, as long as they are adopted from an ethical, playful, and creative stance.

Check out the video and transcript below:

Transcript courtesy of

Leon Furze 0:00
Good. Morning, I’m here with Tom Barrett. And I’m going to let Tom introduce himself because in my experience, Tom does a lot of different things. Today we’re specifically talking about artificial intelligence. But Tom, do you want to give yourself a bit of a general introduction, and then tell us about your interest in artificial intelligence?

Tom Barrett 0:21
Great. Thank you, Leon. Thanks for having me. Yeah, so I’m an education consultant, I run my own business called dialogic learning. I’ve been doing that for about 10 years now, I just realized the other day, it’s been 10 years, I’ve been working as a consultant, supporting schools being involved in a whole range of different leadership and educational product projects. I support teachers and leaders to and with, with coaching. And, yeah, prior to that, I was a primary school teacher in England for about 10 years. So it was an opportunity. Now, I’ve always had a great interest in emerging technologies. So I was a technology coordinator and leader in three different schools that I taught at. And so I have been keeping track of the, you know, technology that is affecting and has the potential to influence our work in learning and teaching and, and so, you know, the artificial intelligence has really started to, you know, the accessibility of these tools has really started to shift. So yeah, I’m really interested in it from a creative angle as much as maybe some of the behind the scenes kind of data kind of angle to.

Leon Furze 1:34
Yeah, and look, certainly for me the creativity aspects of artificial intelligence and education is huge. I’m a methods English and literature teacher and an English and Lit university graduates. And I’ve been really keeping an eye on what’s happening with artificial intelligence around story writing and generation. Last week, I interviewed a guy Mike Sharples over in the UK, who’s recently recently written a book on this called Story machines. And he was really interested in exploring how these writing machines can help us to unpick or unravel the creative process, which is still something of a mystery. You know, this idea that we don’t really know what goes on behind the eyes when people aren’t being creative. What do you think about that? What do you think about how AI might help us to unpick a bit of the mystery around creativity?

Tom Barrett 2:31
Well, I suppose one thing it makes me think about is the visibility of what’s going on with AI when you create, and I think there’s still a way to go with some of these tools and tech knowledge is, and I’m just so you know, I’m, I’m thinking in terms of the technologies, some of the language models that we can use, where we can sort of generate written text. And I’m also thinking about the kind of image generators or art generators that we’ve seen much, much more in, you know, in the mainstream media more recently. So I’m thinking about those sorts of kind of spaces and places where one of the challenges is that black box idea of a lot of the kind of AI technology, and I think even engineers within the field, recognize that that’s the visibility is a challenge. And so I think we have this interesting opportunity that actually increases accessibility to some of this creative process, because I can use these technologies to explore different types of artwork in ways that for me, I can take it in whichever direction I want, which I might not necessarily be able to have been, you know, being able to have done or even not have access to that type of work, not had access to that type of art form, for whatever reason, and we’re able to kind of log on and use these now and have access. So I think there is something, something in there about increasing accessibility to creative methods. But I also think, kind of hidden into that is that aspect of visibility of what actually is going on. So accessibility might go up, but I also think visibility might reduce a little bit. I think that being able to expose the method of what’s happening, and to see process, I think we’re gonna have to start to kind of, I think we’re gonna have to have much better methods and processes to be able to sort of see what’s going on.

Leon Furze 4:29
Yeah, I think that is a really important part of this whole discussion around the ability to be able to do understand the whole process from start to finish. And I think one of the one of the things in education that might rear its head is that teachers are unprepared for these technologies and don’t know what’s going on under the hood. You know, there’s a lot of chatter around the bias in datasets. There’s a lot of talk around how we get the validity of the outputs and AI writing natural languages is notorious for taking itself down little rabbit holes and coming up with sort of its own fantasize tests. So if we look at essay writing, for example, it’ll start off on a prompt 300 words in it’s just completely fabricating what it’s writing about. Up to including making fake references recreating references to journal articles that don’t exist. Yeah. So there’s, there’s a critical awareness of the process, I think that needs to become part of the education narrative around these technologies.

What else do you see is that perhaps the threats and the opportunities of this technology in education?

Tom Barrett 5:47
Yeah, I mean, that point you’re making about prompts. And so if, you know, listeners or viewers are watching this, and you know, not never used these things before, often the prompts prompts is like the starting kind of point. It’s like the instigation of what then what you create. And so they’re often written out early on, so you kind of you write something that then is carried on, and that, you know, you might describe a particular type of text or a paragraph that you want, you may even start writing that paragraph and then ask an AI language model to carry that on. And in, in an art, the art form generate the art generating or image generated is the same, you kind of describe what you want in text form. I think that I think it’s interesting to kind of reflect on that idea of authorship, I suppose as a as potential. It is, I’m not necessarily believe it’s a threat. I think it’s just a shifting definition of what that has is likely to become. And by the way, I mean, this is obviously emerging technologies. And so these are things that aren’t necessarily, we’re not necessarily feeling this yet in the classroom. But I think it’s not far from from us in terms of, you know, I think that students are using, you know, using these sorts of technologies. I don’t think it’s mainstream enough yet. But I think, but that idea, so if I was to frame authorship, is there a threat there? Or is that actually an opportunity? They’re like, can they kind of sit? I’m not sure yet about that idea. But it certainly makes us question, What can I trust? When I look at this piece of work? Like, is this who is the author of this? Who’s created this? It kind of really raises some of those questions, doesn’t it? Because I think that maybe a student or a creator is using different skills is, is using different types of collaboration with technology is expressing themselves in a different way, and is creating in a very unique way with technology. And so authorship isn’t defined in the same way. Does that make some sense to you? Yeah.

Leon Furze 7:58
Yeah, I really like the idea of what you said there about authorship being a threat and an opportunity pretend has the potential to tip either way. And I’ve had a lot of conversations with people recently around, you know, is using these machines, or these machine writers? Is it cheating? Or is it plagiarism? Or is it something new entirely? Is it akin to using Grammarly or a spell checker? Or on the continuum? Can it become like a co authoring or part of the creative process? And is that acceptable? And where do you draw the lines there? And I think that academic institutions, secondary schools, tertiary, and employers are going to struggle with this over the next few years about where that might. Yeah, absolutely, I would really like it to see, to see these technologies being adopted, and all of the ethics and the appropriate use around them being being given to students as a tool that they can, they can keep in the arsenal, rather than something that we we have to limit and kind of push down, which is, unfortunately, what happens.

Tom Barrett 9:02
I remember the days when YouTube was starting to kind of become much more mainstream, and then all the talk about banning YouTube in school, we’re going to ban it, you know, we’re going to ban it, and but the, you know, students that I were working with, were just like walking out the gates and using it within the next half an hour and exploring, you know what I mean? So their lives were kind of a very, very different experience than school. And I think that we have to be mindful of like, what is school if it’s not a place to really critically analyze and be discerning in our choices and critique the technologies that influenced our lives, you know, they should be it should be part and parcel of the of the discourse, it should be a safe space to explore, it should be in terms of a creative process. And we we as educators, I think have a an obligation to understand these creative technologies in order to create such SafeSpace and I’ve been in our schools to be able to talk to our students and learn about the, you know, the depths of them and appreciate what they are appreciate the limitations, and recently read about somebody framing these technologies, as, you know, instrument is instrumentation and you know, is the ability to kind of have another instrument have another tool within and especially for writing, I think it is perhaps, perhaps has the potential effect where it actually creates this new type of instrumentation for for the writing process, where I’m able to rapidly generate, you know, 10s, and 20, you know, hundreds, potentially hundreds of ideas for the work that I’m doing, and make kind of different choices and curate from that. And the writing process, I think is often, you know, different to that is often, you know, it’s often driven by our own creativity and our own processes and our own methods. And the idea of an instrument or something that can you can use to really explore and prototype quickly. You know, through the use of these types of technologies, I’m not sure we’ve seen too much of that. So I think that idea of instrumentation is interesting. And somebody frames some of the artwork generating AI, as giving you a different kind of signature of flavor from like you would with a different instrument, you know, you pick up different instruments to play the same, the same harmony, you experience something differently from these different technologies. And I wonder for writing, whether it has the potential to really be the sort of CO authorship instrument that really helps to kind of explore ideas in a much, much more fluid way.

Leon Furze 11:50
Yeah, no, I like that. I like the music analogy as well, it’s the idea that the AI is going to potentially add a different flavor or a different vibe to somebody’s writing, or to allow somebody who might be a little bit stuck to unlock a little bit of creative capacity. And that, for me, raises potential in a lot of areas from students with backgrounds where they have difficulties accessing technology, or difficulties accessing the curriculum, even this can act as a bit of a bridge, I think, to much more confidence and much more creativity with writing in particular, I’ve played around with a lot of those art generators, as well. And to be honest, they’re just there, they’re a lot of fun. There are a lot of fun to experiment and to play around with. And really, for me, as an English teacher, I want to see writing, and I want to see creating being a fun process. So the tools can be used as part of that, you know, let’s, let’s embrace them, rather than trying to stamp them down. Just projecting forward, you kind of touched on a few of these ideas. But where do you see the future of these tools in education? You know, what does the classroom look like in five or 10 years? Not that we can possibly prognosticate that far, but you know, is this in your mind? Is it a dystopian future or a utopian future? Or somewhere in the middle? What does it look like?

Tom Barrett 13:20
Yeah, I try and sort of steer clear of too much dystopia in my life. But I do have much more hope, you know, than dystopian kind of nightmares. I think, I actually think that when you start to frame it, in a sense that you have a really rich set of creative tools that become much more integrated, then the challenge really in classrooms of the future will be to be to clearly articulate the method that got a student from here to here in terms of outcome, and also to have really clear ways to talk about originality and truth and authenticity in terms of what’s created. And we need much better ways to communicate that so where we are now it may well be, you know, we’re talking about Grammarly. You mentioned, for example, there’s a small percentage of the text that is submitted in a piece of work will have been altered just by running through a Grammarly check. And that’s not. And so what we can do is how do we communicate that percentage, you know, is that like a pie chart that shows that this is my this is there’s, you know, these are the tools that I’ve used. So I can imagine that type of thing being part of this sort of learner kind of profile part. You know, we’re talking a lot about that recently, but within a piece of creative work that’s shared, you know, that it has this a set of clear kind of signals and metrics that show the collaboration that’s gone on, and that’s not to say that, you know, a big chunk or a big part of my pie, I’m saying it is I’ve done with some, you know, this other technology is a bad thing, because then I believe that we’re going to be moving into this sort of phase where new types of skills, new kind of nuanced kind of types of collaboration with technology is going to be something that we have to value as a skill set. So the you know, because, you know, when we look out into the big wide world, you know, we see many, many businesses, marketing agencies using these technologies fully, like there isn’t any barrier, they’re fully using these, you know, they’re, you know, small a marketing agencies engaging in, you know, quite wide scale work using these technologies. And so this isn’t something that is, you know, is emerging in those places, is something that’s directly affecting the output, productivity and creativity of those, those places. So I can imagine our students in the future, being able to, you know, X sort of express in new ways to be creative in new ways, but also the skills that have got them there with these technologies to be something that we value as well.

Leon Furze 16:08
I guess, teaching them how to embrace the technologies and embracing the technologies alongside them, and showing that these skills are valuable and worthwhile out in the real world as well as within education. Yeah. You know, I do love having these conversations, because what I get is a sense of optimism, that really across the board with everybody that I’ve talked to, there’s a there’s a there’s a sense of trepidation, I think, with with education and new technologies, there’s there’s just a baseline and we see that through things like mobile phone bands that we see on YouTube and all of these things. But I think ultimately, in the end, what wins out is, is a bit of common sense. You know, we know these technologies around. We know that it’s our responsibility as educators to ultimately give the students the tools they need to work with these technologies in the best way possible for the best outcomes for everyone. So yeah, look like you. I’m hopeful I’m optimistic.

Tom Barrett 17:07
That’s good, no dystopia for us the

Leon Furze 17:12
Only in dystopia that I teachr as an English teacher, we study a lot of dystopia. We should look at something a little bit happier in the future. Thank you very much for your time this morning, Tom. It’s been great to talk to you again. Hopefully we’ll pick up this conversation at a later date. And we’ll see how we’re going into the future.

Tom Barrett 17:30
Yeah, thanks for having me.

Transcribed by

Leave a Reply

%d bloggers like this: