AI Isn’t Killing Writing. Less Practice Is. | My EdTech Life 378

AI Dread In Writing Instruction

Adam Sparks 0:00

There is legitimate existential dread in K-12 literacy instruction right now, especially when it comes to writing instruction, because of what AI means for writing. And like, does it devalue writing now that a machine can write for us? Does it devalue the the the profession of teaching kids how to write? Are kids going to write? Do they need to learn how to write? Can they just use the AI to do it? There's there's there's so many giant existential scary things happening that I think the predominant emotions right now, with ELA teachers specifically, that we work with is like fear and anxiety.

Welcome Back Adam Sparks

Dr. Alfonso Mendoza 0:42

Hello, everybody, and welcome to another great episode of My Ed Tech Life. Thank you so much for joining us on this wonderful day and wherever it is that you're joining us from around the world. Thank you as always for all of your support. We appreciate all the likes, the shares, the follows. Thank you so much for engaging with our wonderful content and for connecting with our amazing guests. And all of that wouldn't be possible if it wasn't for our wonderful sponsors. So again, a big shout out to Comeback Coffee for keeping us caffeinated and creative all of the time. Big shout out to Book Creator, Eduaid, and Peelback Education. And I am very excited, folks, today because today I am welcoming a wonderful friend. And I really truly consider him a friend from the first time that we met. We shared breakfast together and had a great conversation. And after that, he's been on the show three times prior to this. So I did want to let him know that this is he is one of only two guests, or I guess I should say he's the second guest who has been on the episode four times in the last, you know, couple of years. And I mean, we've done the show for six years. So I really do appreciate him and his insights. So it really does mean a lot when he does get a chance to come back and just share not only what he's doing with his wonderful platform and uh, you know, and bringing, you know, just some amazing tech into our space, but also he shares some wonderful insights and uh, you know, being that he is in the space, he's living it, he's creating. It's so great to hear from somebody like him who's so open and honest and transparent. So I would love to welcome to the show my great friend, Adam Sparks. Adam, how are you doing today?

Adam Sparks 2:27

I'm doing well, man. How are you? I appreciate the generous intro.

Dr. Alfonso Mendoza 2:32

Absolutely, you know, and it's well deserved. Like I said, it's just been great to have been able to connect with you for the last couple of years, seeing what you have been doing, seeing what you have been building. And now, obviously, with a rebrand, which really gets me excited to hear because sometimes it's just those little things that make a big difference. And I'm really excited to hear all the backstory about that and obviously what new uh features are on the platform that are going to help all our teachers and help all our students. But before we dive in into that conversation, Adam, if you don't mind, this might be the very first time, you know, that there we might get uh some listeners that are hearing about you for the very first time. And I'll definitely link the other three episodes that you've been on before so they can get to know you a little bit better. But if you can give us a little brief introduction and what your context is within the ed tech space.

Adam Sparks 3:28

Yeah, for sure. Um, my name's Adam Sparks. I was a teacher for seven years. I taught middle and high school social studies and English in Nebraska, China, and Chicago. So extremely varied and different contexts. And then four years ago, I did my master's full-time at Stanford in their learning design and technology program. Um, we spent a year trying to solve a problem in education using technology and wanted to work on the problem of me having spent a bunch of time teaching humanities classes, giving kids feedback and then having them ignore all the feedback and move on. And what we built at the time, yeah, as a master's project, then as a unexpectedly as a startup was called Short Answer. And then over the last four years, that product has changed dramatically to the point of like, I don't think that name fit us anymore. So in the last couple months, we just we changed our name to Scriblet. And I waded back and forth of should I just wear this hat the whole time, just go over the top on the branding. If you're I guess if you're listening to this, you can't see that I just put on a scriblet hat. But um, but yeah, we're now scriblet. So I'm super excited to share that work um and super excited to to talk about digital technology and classrooms in 2026, because it's a hot button topic all of a sudden, and I'm excited to get into all that.

Dr. Alfonso Mendoza 4:39

Absolutely. But again, before we get into that those hot topics and those spicy takes, those uh, you know, uh, and and I know that's the wonderful part that I know you and I, when we run into each other at conferences and we have the opportunity to talk, you know, it's just great to just hear your insights and those spicy takes. But um let's talk a little bit about just the rebrand. And, you know, now, of course, new name, platform, obviously a little more engaging, more things there. So I want to ask you, I know the last time that you and your wife were on the show, I mean, we talked about how you all went to research and and all of a sudden you just came out there like, oh my gosh, like what's going on here? Like, are are we doing the right thing? Like, you know, for the is our work now part of the problem, you know, kind of deal. But, you know, so that really stuck with me. And but I'm glad that, you know, you guys have really just persevered through this space and obviously with a lot of changes, and then of course the rebrand. So tell us a little bit more now, fast forward to 2026 based on all your experience. Where did that rebrand start? How did it come to mind? Tell us that story because I think that's something that's very interesting for our audience

From Short Answer To Scriblet

Dr. Alfonso Mendoza 5:55

members to hear and learn about.

Adam Sparks 5:57

Yeah, I don't think there was necessarily ever one moment. It was just a series of many different moments over the last four years. Um, when we first launched it, um, we called it short answer for a variety of reasons, but mostly because we thought it was going to be a broad formative assessment platform focused on peer-to-peer models of formative assessment. And that is what it was. But when we put it into hands of teachers, we found that the predominant use case was for writing practice. And so we just kind of followed our teachers down that path of like, okay, you're using this for writing. How can we better support that? And that changed our product roadmap. We started to, you know, instead of prioritizing these features, we started prioritizing these features. And the more we went down that road, the more and more the name short answer was like a misnomer of what we want it to be, to the point that over the last year we worked on developing essentially a supplemental writing curriculum in short answer. Um, and I'm happy to talk more about them. It's a set of features called Skill Builders. And we were like, you know, when we launched this, um the name short answer is not even close to like this, this is no longer like a short constructed response platform. This is now a kind of a comprehensive writing platform. So we wanted a name that was more evocative of that and more evocative of literacy in general, um, because we think the literacy crisis at the moment um deserves more focus and attention in the ed tech space. Because I do think technology does have unique affordances to help solve some of the challenges when it comes to writing, especially in 2026. And so all of that kind of combined, um, there's a million other little instances that led us eventually to um, you know, announcing the name change almost just a month and a half ago at this point to scribble it. And alongside that, that name change, we launched all these new features called skill builders. Um, so increasingly trying to play daily formative assessment of content um off of explicit writing instruction and kind of weave those two things together to create more uh writing practice reps across the curriculum and wanted a name to reflect that, hence the the hat here and and scriblet and um kind of the over-the-top branding that's happening right now to get the message out. Okay, I'm gonna take this hat off. Um, anyway, um I'm happy to go into more detail on any of that.

Dr. Alfonso Mendoza 8:06

Yes, no, absolutely. But I I want to know though, like the name specifically. Who came up with that name? Was that your idea or was that your wife's idea? So that that's something that I think that I would find very interesting.

Adam Sparks 8:20

So it's funny, that was a long story. Um it was it was both of our ideas to answer your question. Um, we listened to a podcast. Um, he's like the naming guru. Uh there's a company that's in the Bay Area. Um, it's called uh Lexicon Branding, and it was with their co-founder. Okay, so Lexicon Branding is the company that named Sonos, which is a speaker brand I love. They named uh the Swiffer Sweeper thing. Uh, they named, oh my gosh, uh Microsoft's Azure platform. Like they're like the naming gurus of Silicon Valley, the type of folks that a small bootstrap startup like us cannot afford. But we can afford to listen to the podcast that he goes on and talks about his process on. And so we listened to his process of like, how do you go about effectively naming a company? And we just followed him down that path. It was a it was a fascinating um interview. I recommend people listen to it, even if you're not interested in branding by any means. Um, his name's David Plasic. It was a fascinating conversation about like um the symbolism of different syllables and the way words sound and how to go about standing out in a space. And so we followed that process. And what we eventually landed on was a combination of the words scribe and scribble, because we liked what both of those concepts denote, um, mixed with lit, L-I-T, so scriblet, um, because we liked the internal alliteration of it, like the rhyming of scriblet's kind of fun to say, it's easy to say, it's only two syllables. Um, kind of sounds like, I mean, it's a noun, but it's very verby and it's also made up. So we get to imbue it with our own meaning over time. One thing that we didn't like about short answer was very constraining. It's like everyone has preconceived notions about what a short answer is. Um, and we got all sorts of bad jokes in response to it when we were short answered. Um, so we like the idea of a made-up term. Um, we we like the idea um it's scriblit L-I-T, but when you say it out loud, it kind of sounds like scriblet L-E-T. And the suffix let means small. And we wanted to denote small, simple, easy, um, because I do think if you look at the K-12 writing space at the moment, a lot of the platforms are the opposite of that. They're these big, heavy, um, you know, it takes an hour-long professional development session just to learn how to click all the buttons type of platforms. Um, and they're not necessarily bad pedagogically, but from a user experience perspective, I think they leave a lot to be desired. So we wanted a name that was more evocative of that. Um, I mean, I could go on and on, but that but to answer your question, like it was both my wife and I, and that's partly why we wanted to do this. Short answer was kind of my thing. That was my master's project. But now, um, you know, I've been building this with Alexa for the last four years. We wanted it to be our thing. And so that was part of it too, for sure.

Dr. Alfonso Mendoza 11:03

No, and I, you know, I love it, you know, and that's such a great explanation, Adam. And, you know, and I'm glad that you shared that because for our audience members that are listening, you know, having that context and and seeing how you, your wife, you know, as as you continue to build, you know, there's nothing wrong with a rebrand. And it's really just still taking that central message of what you're trying to do, which is to help support writing. And we'll get into that as far as the updates and the skill builders. But I think that this is something that's very important to be able to make that connection to your platform, making that connection to your users, your teachers, your students. And I absolutely love the name. It just rolls off great. And, you know, it just for me, you know, I I'm curious and I want to ask you because, you know, with a with a name like Scriblet, you know, educators can kind of get an idea of maybe what the platform will do. But I want to ask you, you know, within these next three years, what do you hope that Scriblet means for teachers and students?

Adam Sparks 12:10

It's a good question. You mean like when they hear the name, what do they think?

Dr. Alfonso Mendoza 12:15

Yeah. I mean, with the name, what they think, what they associate it with, like, what

Making Writing Feel Joyful Again

Dr. Alfonso Mendoza 12:20

do you hope that, you know, in the next three years, maybe five years from now, too, you know, when they hear scribble it, it's like, bam. Like, what do you what do you hope that teachers mean when they say it?

Adam Sparks 12:33

Yeah, that's a really good question. And I don't know if it's a thought. I think it's an emotion. And I I just I hope they smile when they hear it. And the reason I say that is there is legitimate existential dread in K-12 literacy instruction right now, um, especially when it comes to writing instruction, because of what AI means for writing. And like, does it devalue writing now that a machine can write for us? Um, does it devalue the profession of teaching kids how to write? Does it um are kids going to write? Do they need to learn how to write? Can they just use the AI to do it? Like, there's there's there's so many giant existential, scary things happening that I think the predominant emotions right now, with ELA teachers specifically that we work with is like fear and anxiety. And that is a mega bummer because I don't think anybody went into education to teach because they wanted to ensconce themselves in fear and anxiety. They wanted to ex, they wanted to do it to to I mean, imbue joy and hope and and love and and take care of kids and and and prepare them for the world. Um, and I don't know, there's just a lot of fear and anxiety right now in K-12 writing instruction. And so if and when they hear the name scribble over the next five years, I hope that it it elicits the opposite reaction. I hope it elicits smiles because it's creating experiences that help kids fall in love with writing. And that's that might sound very um CEO-y kind of you know, vibey inspirational, but I legitimately mean that. And that's partly why, and I mean this sincerely, that's why we changed the name. When we were, we had we came up with a bunch of different names, which I won't bore you with here, but we were leading a session this summer in Abilene, Texas, at uh region 15 at their um put on a great conference. I can't remember the name of it now. Um but it's kind of an ed tech conference. And we were presenting in a hallway with 15 or so teachers, and we had some time at the end of the presentation. And I said, Hey, we're currently called short answer, but we're thinking about the name change. What can I can I kind of float some ideas by? And so I just pulled up a slide and I started typing it. Um, and once I got done with it, nobody said anything, um, but they all smiled. And that is exactly what we wanted. Um, and so that was kind of the moment I was like, okay, this might be the one. Um, because that's the emotion we want to elicit. And that's kind of the experience we want to create both for kids and for teachers. So I don't know if that I don't know if that answers your question, but that's that's the emotion we want to elicit.

Dr. Alfonso Mendoza 15:01

No, and it actually does. And and even just uh, you know, to be honest with you, and uh, I mean, a lot of people may be like, well, it's because he's a four-time guest and everything, but it it it's just because I've I've seen what you have done, you and your wife both, the work that you've put in, and really the heart and the passion that you guys have put into this platform, into the work. And it truly like it it doesn't sound CEO-y, it doesn't sound very vibey, it just sounds really transparent, you know, what it is that you're trying to convey. And even myself, in my mind right now, as you're answering, and I just continually say, scribble it, scribble it, like I just can't help but smile either, you know. But but I love the way that you you put this in in the sense for the teachers, you know, we don't platforms, you know, many times they already come in and they seem very ominous. Adding something else to a teacher's plate seems very ominous. Right now in Texas with state testing and writing, and and I and I'm only saying state of Texas just because this is where where I live and I reside, and I know the way that test goes, I know the levels of anxiety when I see a third grader come in and immediately walk out five minutes after the test starts, crying and feeling nauseous and just that anxiety kicking in. I mean, that's not fun. That's not why we become teachers, like you said, that to not only anxiety for ourselves because of state testing, but to pour that anxiety onto students as well, you know, because we need to get these scores, we need to get those A ratings and so on and so forth. So I I really do love what you mentioned and it really does. Like it just rolls off the tongue very good, scribble it and yeah, it just nothing, it forms a smile on your face as you're saying it. So, but now I do want to put my former assessment

Authentic Writing Evidence In Class

Dr. Alfonso Mendoza 16:55

coordinator hat on here for a second. And I so I know that one of the reasons for the change also is because AI has made assessment validity, validity or validity of the assessments a little bit harder. So now you've moved the writing back into a classroom, having live sessions and so on. So, as somebody who formerly worked with assessments, I want to ask you is the in-class that visible writing the new baseline for authentic evidence of learning?

Adam Sparks 17:32

Um, I don't I I don't want to say yes because I I don't want to over-communicate or like or communicate that all assessment has to take place in class because I don't think that's realistic. But I do think going forward, the overwhelming majority of work going in the gradebook should be stuff that was completed in class. Um, now, whether or not that relates to our work at Scriblet, it's tangential because um we don't really intend a lot of the stuff in Scriblit to be put in the gradebook. Ours is more for low-stakes formative and formative assessment and instruction purposes. Um, but I do think going forward, yes, more assessment needs to happen in the classroom. But that's particularly hard with writing because writing is longer form in nature and not something, you know, if a kid's writing a five-page research paper, they can't just sit in class and for a week straight and do nothing but writing. Like we have to take, you know, what valuable in-class time we have and and teach our standards and and cover content. So um it's created a unique challenge for writing, I think, because of the more asynchronous and time-consuming nature of it. Um, so I do think we need to bring more writing into class. Um, but I'm not saying all writing needs to happen there. Um and so that's partly why we structured, scribbled around short form writing. It's intended to take place in the classroom. Um, but it's also why we're we're trying to embed more writing across the curriculum. Um, because I think going forward, kids are gonna get less practice with writing because they are gonna use it, uh, use AI in their in their personal lives and in their digital technology platforms that they use on a daily basis. And I think the more that they do that, the less they're gonna write and the more your skills attribute, the less you practice something. So to make up for those lost practice reps, uh another part of what we're trying to do with Scriblet is create more practice reps to make up for them by embedding it across the curriculum and doing it in a way to you to a point you mentioned earlier of like where it's not one more thing for a teacher, where we're combining daily formative assessment of content that they're already doing with targeted explicit writing instruction. And that's hints the the weaving together of what previously was short answer with this new set of features called skill builders. So uh we're hoping that that's a more holistic approach to um adjusting the way that we approach writing instruction in K-12 schools to uh to address some of the challenges created by AI. And like it's a silver bullet, but that's definitely a first step.

Dr. Alfonso Mendoza 19:42

Hey, you know what? And first steps, that that's that's the way to go. But I appreciate that very much. And one thing that you mentioned that I do want to, you know, just highlight is the the the different curricular writing. I I remember being in the classroom the last couple of years before I transitioned to digital learning coordinator, uh, we would do what we would call our whack writing, WA writing across curriculum. And that was something that we needed to do. And so to gain practice. And man, I remember how interesting that was. And even to this day, I was having a conversation with a colleague earlier today as well, and talking about how even though when we might be out there at a conference and showing either a platform or something that can be done, even within teachers, you say, Oh, well, that's just for that's just for writing. I don't teach writing, you know, I'm a math teacher. Well, I mean, I know that there's writing in math, you know, but it's almost like we we think in silos. And even our students, and my experience was being a science teacher, social studies, and math teacher, they're like, well, I mean, it's it's science. Like, what am I gonna write about science? It's not writing class. So I I think that that is something that is very important. So I love what you're doing there, bringing in the writing from different contents. Students can see, like, hey, the writing is not just siloed for ELA, for test taking, but really I'm writing it in all content levels. And I think that's something that's very important that we do need to practice more, which brings me to the next question. I know you're talking about the skills builders. So I want to talk to you about, or the skill builders, I should say, that are in Scriblit. And so tell me a little bit about that, you know, going from previous platforms, short answer, and how did you come up with the idea of, you know, having these skill builders? What were some of the things that you were already seeing through formerly known as short answer that you said, hey, we need to build on this to, you know, continue to grow and make that writing accessible for all teachers and students. So, Adam, this brings me to a next great point. Like I was talking

Skill Builders And AI Feedback

Dr. Alfonso Mendoza 21:55

to you about, you know, writing and writing across curriculum and working with students. So I want to ask. You, the idea of skill builders, how did that idea come about? Understanding that obviously coming in from short answer and now changing to scriblet, what were you noticing? What were some things that might have been lacking as far as short answer is concerned? But now you're saying, hey, let's now that we're scriblet, this is what we want to do, this is what we want to build on. Yeah.

Adam Sparks 22:24

I mean, the biggest flaw was like our teachers were telling us this is a fantastic platform. And when I say that, I'm referring to short answer, like this is a fantastic platform for practicing writing, but it's not necessarily teaching my kids how to write better. You know, I'm still, it's still dependent on me and my curriculum to do that. And I that they were wanting more instruction on like, okay, yes, this is a great platform for talking about what's a high quality, I don't know, thesis statement or uh, you know, et cetera, et cetera. Name your favorite writing skill. Like it's it's a great platform for practicing and discussing, uh, but not necessarily for explicitly teaching. And so we were hearing that. Uh, and at the same time, we were looking around the landscape, and it's like, there, there are other digital platforms that do have this sort of um um like targeted explicit writing instruction. Um, so some of the more popular ones are called like no-reading, uh, writable, quill. Um, but we didn't feel like a lot of those platforms were tapping into the new unique affordances of AI in the ways that they could be. Um, a lot of those platforms were relying on kind of selected response-based tasks for students as opposed to like open-ended writing tasks. And they were doing that because historically you had to rely on some of the scoring of these tasks. Uh, if you're going to do it digitally at scale, you had to do something that a machine could score. And so it had to be selected response tasks, like multiple choice or click the sentence multiple times to change the, you know, the, let's say, the grammar or whatever structure you're practicing and then submit it. And then we're like, well, there's some cool things you could do with AI to like augment that and approve it and the way it's being used in the schools. And so, like, well, what if we kind of combine our open-ended formative assessment with that model of the explicit kind of sentence paragraph level um writing skill instruction? And so that's that's kind of what we've done. So, like with our um open-ended writing activities, say a battle royale where the students are submitting their responses to a to a prompt that the teacher creates, and then selecting the top response in the class and then reflecting on that top response. That remains the same. But then on the back end, what we're doing is analyzing all students' submissions and suggesting skills they need to work on based on those submissions. So it's like, hey, it looks like in this class of 20 students, eight of your students had sentence fragments showing up in their responses. Um, maybe that's a skill that you should work on. And there are a lot of platforms that will tell you that today of like, hey, um, here's the general analysis of how your class did. But I think we wanted to take it one further and be like, and in addition to how they did, here is what you should do, and here's exactly what you need to do it. So you can, it's like, okay, sentence fragments, you click a button that says generate skill builder. Um, it's gonna generate an activity based on the content that you were teaching. So let's say your prompt was about the American Revolution. It's gonna generate an activity where the kids are practicing turning fragments into complete sentences, but it's doing that all embedded in the content of the American Revolution because AI was used to generate it. So it's keeping them embedded in the content, but still practicing the explicit skill, which we think is a valuable way to make it easier to embed these types of practices across the curriculum and it embedded in teacher workflows. Um, and then also, again, you can share this out just like you do any other scribble activity where you share a code, kids sign in. And then because the kids are doing it digitally, it's being scored digitally. Um, but because we're using AI to do the scoring, students can also write whatever sentence they want. They're not limited to a selected response task. So they can write in more authentic ways, and then it can provide rich feedback on their submissions on those different skill builder tasks. So there was just a lot of unique affordances, and we felt like it was a great opportunity to tap into some of those affordances of AI to improve the way that writing instruction is happening and kind of combine it with formative assessment to address some of those larger challenges being presented by AI. And it took us, I mean, we developed this over the course of a year. Um, I I will tell you, I massively underestimated how much work was going to the end into this. When we first conceptualized this, it was like, you know, we should be rolling out a pilot of this with schools in like six months. And it was like, yeah, a year later, we're rolling out the first version of it to our to our schools. So um, and that was just a month and a half ago now. So it's it's been a long time coming, but I'm really proud of it. And the early feedback from skill builders has been super positive. So very excited about um what we're seeing thus far and and kind of where we're going with that.

Dr. Alfonso Mendoza 26:42

I love that. Uh, and that's great. And that's so great to hear. And a couple of things that I know, like from getting to know you in the very first time, you know, that we talked, you knew me right off the bat. I was always I was kind of just that cautious advocate. And so again, now putting a different hat on. I put my assessment hat on. Now I've got my cautious advocate hat on. So I know you mentioned with skill builders, students are writing. They're, you know, the goal is to get them to write more authentically. And of course, the AI is what is grading, guiding students. But for other cautious advocates like myself that might be listening, can you walk us through a little bit of your some of the guardrails that you do have uh built in? Obviously, because as you know, when you do give students that freedom to write, and maybe that authentic writing can be a little too authentic for school in that sense. So can you tell us a little bit about that? And then my question to you is also um,

Guardrails Plus Where AI Stops

Dr. Alfonso Mendoza 27:40

where should AI never be within writing instruction?

Adam Sparks 27:47

Oh, wow. Those are two big questions. Um, in response to the first one, first and foremost, we do have a content filter. So so kids can't submit naughty words. And if they try to, it'll say, like, hey now, you need to get rid of these words before you submit. That goes, that's been in our app for a while. That's not new with skill building. Um, in addition to that, you're asking about kind of AI guardrails. How do we keep this safe? Um, students never interact with AI like in a back and forth sort of chat bot interface. Um, and students are never using AI to do the task for them in any way. It's the students are always doing all their writing themselves. Uh, and then the AI is coming in afterwards to either give them feedback. And in the context of a discrete grammar skill, that feedback is literally just like, yes, this was right, no, this was not right, and here's why. Um, and so it's very there, there is no interaction there. I mean, we do have one button where you can click why and it'll pop down a pre-populated explanation of based on what you submitted of why it was correct or not. Um, but there is no interaction with it. So it's we we can keep it really constrained to the goals of the activity. Um, same goes for our AI feedback in the open-ended writing activities. It's only going to provide feedback focused on whatever the teacher set as a success criteria. Um, and then even within that, there is no interface back and forth with the AI. Uh, and students can flag it if they disagree with it or they find it unhelpful. And it's like, hey, we noticed you flagged your feedback. Do you want to discuss it with the teacher? And it'll flag it on the teacher as well. And that's kind of like the UI kind of like user experience ways that we're putting guardrails and thinking through how people might interact with AI in um in productive and meaningful ways. On the back end, we can get into a whole host of other things if you want to. I mean, we we we um offer our AI services through um AW Amazon Web Services Bedrock platform. Um Bedrock has a lot of advantages. Number one being we can pull from a variety of different models that might be better or worse at different tasks. So we're not beholden to like only using anthropic and cloud or only using OpenAI and Chat GPT. Um, but then also bedrock has the has a great affordance of essentially making a copy of models that you can put a walled fence around such that you control all data that is or is not being shared with model providers, which in our case is we provide we share no data with um with model providers, aside from the responses with the model like directly, but it's in a walled guard so that the data that's being shared is not being used to train the models or anything like that. Um so AWS bedrock is is has been huge for us and how we go about structuring it safely on the back end. Um, I mean, we also are happy to sign data privacy agreements with school districts, um, and all those DPAs take care of, you know, what we're not using any data to to market and sell things. We're not selling data, you know, we're following Copper Furpalas, all that good stuff. So we could go into the back end kind of technical legal side of it, but I think your question was more oriented to like the user experience side of it. I kind of hope we've addressed that. And then your second question was um say it again. Is that true?

Dr. Alfonso Mendoza 30:51

My second question was like, uh, where should AI, or you think, where should where do you think that AI should not be included or be participant of in the writing process?

Adam Sparks 31:03

Man, it's hard to answer that at a blanket, you know, just like a blanket statement response because it's very contextual. But generally speaking, the way that I think of it is like a stoplight system. And I know a lot of people have problems with the stoplight classroom management system for AI. We can get into that if we want. For those that aren't familiar with that, it's like if I label the assignment red, they can't use AI. If if I label it yellow, they can use some AI with constraints. If I label it green, they can I want them to use AI any way they want. That's partly what we're measuring in this task. Um, that's partly the way I think of it, um, but in a little bit of a different sense. So if a student does not know the content or the skill, I think the default with AI should be red. Like the kid shouldn't be using AI, because I need to know what the kid can and can't do on their own. Because I do think to some extent, effectively using AI is predicated on being an expert in whatever you're using it for. So I think uh that would be my my default would be red, like if the kid doesn't know the content or the skill. But that said, um, that's that's if this kid is using the AI to do the task for them. I think there's an interesting gray area when it becomes like how might I scaffold the instruction for that student during learning to support them, but then still have them do the assessment task on their own. And so I do think there's still a place for AI in there, but not necessarily to do the task, but maybe to scaffold the task. So I'll give you an example. So in Scriblet, when the kids are writing, we have the the teacher can use AI to generate sentence stems for the student. Um and the kids can then use those sentence stems if they struggle with getting started with writing to plug them in and they can basically finish the sentence to help them to help them get started with writing. Um so in that way, they're technically, I guess, using AI, um, but they're not using it to do the task necessarily. It's just to scaffold the task. So that's where it gets very gray. Um, so I guess it's like most assignments should be red. They're not using AI to do the task, but you might be yellow in the instruction portion of it. And then I only really think AI should be green, as in like full blast, you can use AI however you want. Um, if and when students already have mastery of whatever you're teaching them and and what they're what you're doing with the assignment is measuring how well they can use AI to augment that mastery, if that makes sense. Um, so that it's sort of that red, yellow, green helps me scaffold the way that I think about the yes-no question when it comes to AI, but it is deeply contextual, dependent on age, dependent on expertise, dependent on what your assessment construct is, like so much that it I don't think there's any blanket labels that can be made. And I think that's partly what makes some of these AI bands kind of misguided, because they um they don't allow for that, for that gray area and wiggle room. It's it's it's all it's either black or white, it's good or bad. And I don't necessarily think that's realistic.

Dr. Alfonso Mendoza 33:58

Yeah, no, and especially that that last comment that you mentioned, it's it's really it's either yes or no. And to me, in my mind, the very first thing that comes to mind, and again, just because of my experience growing up, an English language learner. So I immediately I think of my multilingual learning students. Like, what are we doing? How are we gonna support them? Then I'm also thinking of my students that receive additional services, supports, special education. What are we doing for them? Because it's either one or the other. And either way, it's like, okay, if it's allowed, great, those students are gonna be able to have access. But then if it's completely banned, then it's like, okay, how are we gonna support the students? And I know people can say, and I saw somebody post this, um, and I can't remember who, but I remember seeing, like, hey, well, we used to do it without tech at some point in time. But yes, but do you remember how hard it was when you did have those 27 students and maybe you needed those five students that needed those additional supports? How much time did you really allot for those students that they did not so they wouldn't fall by the wayside? And now with technology being available, that students can be able to access or have greater access to the learning through translation features or um, you know, just uh a tool that helps explain or helps them guide them in the writing. I mean, you're seeing that growth and that support. So kind of brings the conversation now into kind of the tech backlash. And I know we kind of talked a little bit about pre uh a little bit about this pre-chat.

Screen Backlash And EdTech Accountability

Dr. Alfonso Mendoza 35:35

So I want to ask you uh your thoughts on this, because we'll we talked a little bit about this. So when you have uh the the a parent come to a school leader saying, hey, kids need they they need less screens, less screen time, there's too much screen time, you know, where do you think that they might have a point, if any? And what might be, you know, something that you would say, well, I I don't think that they're thinking this through completely. So what what are your thoughts on that with everything that you've seen?

Adam Sparks 36:12

Well, if if an angry parent is showing up in my office as a principal or administrator, I think my first response needs to be empathy. Like, I don't think just rejecting it out of hand is going to be productive for anyone involved. And I do think there's legitimate things to be empathetic about. Like parents are legitimately concerned about the digital world that their kids are navigating. Uh, and they have every right to be because there's a lot of scary technology out there that has been proven to be manipulative of kids. I'm talking about a lot of social media and ways the features have been set up in social media and targeted advertising. Uh, you know, name your favorite like Facebook, Instagram, TikTok scandal when it comes to like students interacting with technology. Um, I mean, there's there's studies showing that kids are spending anywhere seven to eight hours a day on a screen outside of school. Um, I totally understand being very anxious about that. Um, but then I would quickly follow that if I'm an administrator with we need to separate what's going on on a cell phone outside of school with what's going on on a managed device under the guidance of a licensed professional in a classroom. Because they're two very different use cases. And um there are many positive affordances of technology that we shouldn't just throw out the window because we're worried about pick your favorite topic that you see, you know, a Twitter headline on. Um and so that would be my response is I I would first try to get the parent to see the difference between personal use of tech and commercial tech and educational technology used under the guidance of a professional educator. Um, and then if they wanted to, we could go into the specific use cases and affordances that that make technology valuable. You've already talked about some of them. I think first and foremost, it's differentiation of instruction for students with disabilities and with IEPs and behavioral intervention plans. Name your favorite thing that you need to differentiate for. Like technology is critical for that. I worked with a teacher in um Chicoche, Oklahoma, who uses a tool called um Microsoft Immersive Reader every day because she works with students with dyslexia, and immersive reader allows you to take the text and make it much larger. You can break it into syllables, you can hover words, and it'll give you an image of the of the concept. So you have that uh that image representation, and it'll it'll show you the adjectives, nouns, verbs in a sentence. It'll read it, it'll translate it to different languages. Like there's so many amazing affordances of a tool like Microsoft Immersive Reader for differentiation. The idea that we should limit that teacher, you only get X minutes per day to use that because screens are inherently bad. Um, I think that's absurd. I don't think the research supports that. I think that's more of an emotional decision than a than uh than one based on evidence. And I think decisions in school should be based on evidence. Um, so that's one use case in affordance. We're talking about differentiating instruction for students with dyslexia. There are many other contexts in which we could talk about why technology is valuable. Um but so anyway, I uh that's how I would respond to a parent. I think you had another question in there, which was um where is that where are they right? Or like where is some of the pushback valid?

Dr. Alfonso Mendoza 39:15

Yeah, where would some of that pushback be valid? Yeah, that's correct. And especially now, I mean, we're seeing what New York is doing, we're seeing what LAUSD is doing, and there'll probably be some more that'll be slowly maybe following suit based on what they're seeing. So, yes, there there is this side that you discussed, but what might be some positives in that, I would say. You know, I know and I know it's very hard for me to say, like, okay, well, this is a positive thing because you know, my mind always goes to my, you know, under underserved student populations that really need those like those supports. So what might be something good that may come from this, if any?

Adam Sparks 39:56

Yeah, I think first and foremost is it's reorienting how we evaluate educational technology tools towards evidence and away from engagement metrics. Um, I think that's a very good thing. Um, like the fact that we're yeah, we're we're getting into the nitty-gritty of implementation of what are the positive use cases and what are the negative use cases, and how do we tap into the good and mitigate the bad? Those weren't conversations that were happening even just like a year or two ago. So I I definitely don't want to, I don't think this is all bad. Um, if we can orient the backlash towards productive reform, but unfortunately, I'm not seeing that. What I'm seeing is what we always do in American policy reform, which is we swing the pendulum as hard as we possibly can in one direction, which is ban it all. It's evil, you know, angry parents showing up at board meetings, screaming and and like literally and and upset and booing, and um, and and and they've been worked into a frenzy by people who I don't think have their best interest in mind. I'm happy to talk about Jared CUNY Horvath and Jonathan Haidt and other folks that are selling books and consulting services on the back of this backlash. And uh I've obviously have strong opinions on that and disagree with a lot of what they're arguing and don't think it's supported by evidence. But I do think one positive of a few here that we could get into is that yeah, it's reorienting ed tech evaluation towards evidence. You see that at ISTE and some of the things they're doing. You see it in a lot of the policy reforms happening in a district level. So I think that's a good thing. And I think that's something we at Scriblet are well positioned to tap into because we've always leaned in. I mean, this was my master's project. We we we created a theory, research back theory of change before we even built the tool, which is, in my opinion, how ed tech should be built. And oftentimes that's not how ed tech's being built. It's being built around, well, let's build a tool that's really addictive and fun to use, and then we'll find a way to retrofit some research to support what we did because we know it's gonna be popular because kids really dig it. That's always, I think, been backwards. And I'm hoping this moment can can get us to flip that around. And so we're always starting from a place of, well, what does research and evidence say first? And then we'll figure out how to make it an engaging user experience and make it something kids love. Um, because I do think for a long time, and for for most of my life when I was teaching, um, and even you know, early days of developing short answer now, scribble it, the predominant way that people are choosing or not choosing to use technology in their classroom is based on just word of mouth and like, do the kids like it? Um, and there's nothing wrong with word of mouth and do the kids like it, but it certainly shouldn't be the only reason we're choosing to put technology in front of a kids. Like the first and foremost should be like, does it drive outcomes or not? And I do, and I do think this movement is is hopefully pushing us in that direction. So that can maybe be one positive.

Dr. Alfonso Mendoza 42:32

Yeah, no, no, no. Yeah, I agree with you, you know, I think it was just because and you brought it up, and you knew from the very beginning when I talked to you, it was like, oh, they gave me a shirt, so I'm gonna use the platform. Well, they gave me some shiny stickers, you know, and some of this, to be honest with you, I think was brought on by the industry itself, you know, in the sense of too many tools, too many logins, all the shiny stuff, all the swag, you know, all of that, you know, shiny stickers and everything. And so now I want to ask you, you know, as a founder yourself, if you would have the opportunity to have maybe 10, 20 founders right in front of you in one room, what would you tell them that they themselves or their, I mean, as a whole, uh the ed tech industry needs to own up to?

Adam Sparks 43:29

Well, first, I would buy everyone a beer or an equally attractive non-alcoholic beverage, and I would pat everybody on the back and say, I'm sorry for how dang hard this is because I know it's not easy. And that's not even in reference to the backlash. That's just in reference to how hard it is to build uh a successful startup. I mean, it was already hard to do this. K 12 was already really hard, and now on top of it, you're being treated like you work in the tobacco industry or something like it's hard. So I I want to start with empathy. I'm always gonna err on that side. But then um, to your question. More forthright response to your question of like, what do you know, if if if you made me the dictator of ed tech or if I was in charge, what would I tell fellow founders? Um man, that's a good question. I guess it would be kind of what I taped on or what I talked about just now of like, is there evidence to support your tool or not? And if not, why not? And like legitimately look yourself in the mirror and ask that question. Like, why are you building what you're building? Um, because I do think the overwhelming majority of people I interact with are doing it for the right reasons. Like, I I've interacted with we always try to put our booth next to the magic school booth at conferences because because magic school has a lot more resources than we do. And so they bring a lot of foot traffic. And so we're we're like the barnacle on the side of their ship trying to pull people into our little booth. And so, all that to say, I interact with a lot of the magic school people. And almost all of them are former educators, including their founder. And when you interact with them, it's like, oh yeah, you guys are doing this for all the right reasons. Like, this is awesome. Like this is where it should be. And this, and you're doing this because of lived experience of your time as an administrator, as a teacher, whatever. Um, I don't know if that's everybody in EdTech, though. You know, I I've interacted with other folks that are may not be coming from that perspective. Let's put it that way. I don't want to be too negative. And so I think, yeah, like look in the mirror and like, is this growing out of a need or is this growing out of a desire to make money? Um, and I, you know, sometimes those two things are inseparable. It's hard to draw a line between them, but um I don't know. There has been some interactions I've had with folks that have rubbed me the wrong way in that in that sense. Um, and I would I would just love to see more of an orientation towards impact and less of an orientation towards scale and growth and sales and and the next funding round and you know, impressive LinkedIn posts. Um and so I guess I don't I don't know if that's one thing that would just be more like uh general direction I would point us. I don't I sometimes say this when I lead workshops, I don't have a map that's gonna tell fellow founders directly where we should go. Um, but I like to think I have a compass that at least points us in the right direction. And I think that compass should be impact as opposed to sales and scale. And I think sometimes scale and sales is the comp is the direction we we point ourselves as opposed to impact. Um yeah, that would be my right to my fellow founders in that environment.

Dr. Alfonso Mendoza 46:31

I love that. I love that, and it's wonderful advice. And like I said, truly, like, and for our audience members, you know, listening to Adam, I mean, this is the way if you've never had the opportunity to meet Adam in person and talk to him, like this is the way he really is. This is, I mean, he's just honest, very sincere, very genuine guy. And he'll just let you know like how it is. And sometimes he'll share a spicy take or two and things of that sort. But I mean, I really do appreciate your answer there. And, you know, it's very interesting because I know I've had a couple of guests on the show, uh done, you know, researchers, and they'll say, you know, obviously logins does not mean engagement. So how are we measuring that usage? How are we measuring the efficacy of the platforms? And as a matter of fact, uh, you know, the recent episode that I released with Daisha Toll, who uh works for Course Mojo, you know, she said here, and and I'm just reading the right this right off her transcript right now, which she states historically as ed tech, we have gotten paid not only regardless of whether uh there out the outcomes are achieved, but regardless of whether or not the product was used in a lot of cases. So I mean, she's kind of just in a way calling that out in some ways of hey, maybe we want outcomes-based contracts now, something to look at, you know, because now you have to show the research. You have to pull those numbers. And one thing that I do want to say, and this might be a very spicy take, you know, if you're posting on LinkedIn a beautiful, colorful PDF and, you know, with numbers on there that look very nice, I always ask, can I see the research? I want to see the actual research. Um, you know, being a researcher myself, because I want to know how you come up with those studies too. And so just be transparent, you know, you don't, you know, just be honest with what you're putting out there and then just do those studies, put in that time and then just continue to move forward. Because I think, like you said, you know, with this tech backlash, those that that thought process, now it's that reorientation of okay, how are we gonna go and navigate this? How are we gonna adapt? Now districts are gonna want to look for studies for data. They're gonna want to use, you know, if we've done any pilots. Now we have to show and we have to show our cards. We can't just get by with shiny stickers, wonderful mascots, shirts, and having the teachers just plug away at our platform within their classrooms on their own, you know. So it's very different now. Very interesting times that we live in, Adam. But I'm thankful for you. No, I'm thankful for you, the work that you're doing. And again, just the vision that you have. And I know that we kind of got into this, but I'm very proud of what the work that you've done, you know, bringing up from coming up from short answer now to scriblet, uh, bringing in the platform, helping our students become better writers and being a wonderful resource to help supplement the learning and help our students continue to grow. So I really appreciate that work. And before we wrap up, I just kind of want to ask you just one last question here. And this is kind of maybe like a full circle question for you. So, what I want to do is I want to take you back to where it all started. So, again, I want to ask Adam, the classroom teacher, you are grading papers all weekend. You're, you know, watching kids and ignoring your feedback. All right. So if that version of you could see Scribble It today, what do you think he'd say? And what would you want to tell him?

Adam Sparks 50:13

What do you think he'd say and what do you want to tell him? I don't know. I I um I think I think Adam, you know, circa 2018 teaching eighth grade social studies in Milford, Nebraska. Uh, if you told him that he was gonna go to Stanford and build a net tech startup that supports over 100,000 teachers and students, he'd be pretty ecstatic. Um that's pretty cool. Uh and so I so I'm intensely proud of that. Um at the same time, as much as possible, I'm I I want to make sure that like Adam the teacher isn't something that lives in the past, you know. I I think that can be something that's hard about building ed tech tools, it pulls you out of the classroom and it you can kind of lose touch with reality very quickly. So I've been trying to get into classrooms more. I went and co-taught a lesson last week in uh Bennington, Nebraska with my cousin, taught an eighth grade social studies lesson. So I'm still trying to put on my Mr. Sparks pants every now and again so that I don't, you know, have to live in nostalgia land about what was it like to teach. They know it's like what you know, let's teach right now. Um and so I think the more that we build scribble, the more I want to do that sort of coda design and co-teaching with teachers because that's incredibly valuable feedback and keeps me grounded in the reality of classrooms. Um so I I don't know, I kind of cheated, but that's that's what I think. I I would think Adam in the classroom would like to think of Adam now, but also I want to continue to be Adam in the classroom.

Dr. Alfonso Mendoza 51:33

I love it. Great, great, great, great answer, man. Well, Adam, thank you so much. I really appreciate you. Thank you so much for being here. And again, an honor to have you on the show as a fourth-time guest. Not a lot of people get to do that, and you're one of two that does it, but it's always great to have you here and sharing your insights, sharing your work, and obviously your your perspective on this, not only as teacher Adam or Adam the teacher, but as Adam founder, co-founder. So that's wonderful. And you never lose sight and perspective of what it is, what it is that you're doing. So congratulations on the rebrand, congratulations on the work that you're doing, man. And I look forward to seeing some great things uh from Scriblet in these next couple of years. So just keep at it and we'll see when we'll have you back for the fifth time. So we'll see. But I appreciate you. And uh so, but before we wrap up again, last thing, if you can tell our audience members that are listening, they're like, hey man, I just really want to connect with Adam. He's a great guy. I want to learn more about Scriblet. What is the best way that they could connect with you?

Adam Sparks 52:38

Yeah, just email me at adam at scriblet.com is the best way. I mean, we have all the social medias. You can follow us there, but adam at scriblet.com is is the best email to reach me at.

Dr. Alfonso Mendoza 52:46

Perfect. Well, definitely make sure that we post that in the show notes. And for our audience members, please make sure you visit our website at myedtech.life where you can check out this amazing episode and the other 376 episodes, I believe, by the time this one releases. And I promise you, you're gonna find some knowledge nuggets that you can sprinkle onto what you are already doing. Great. So thank you again for all of your support. Please make sure that you follow us on all socials at myed tech life. If you go to YouTube, please make sure that you click on the little bell and make sure that you follow us so that way you can catch all the wonderful episodes and please make sure to share our content, guys. We want all the wonderful algorithms to get us out there to more ears. But thank you as always for all of your support. And as always, my friends, until next time, don't forget, stay techie.