Latest / Educating All Learners Alliance / Balancing Innovation and Guardrails in AI: A Candid Conversation with Erin Mote
Transcript
- 0:00Hi, this is Aurora from the Educating All Learners Alliance,
- 0:03and today we're speaking to Erin Mote for a deep dive into prioritizing students
- 0:07with disabilities and AI policy.
- 0:10In April 2026, EALA released version 2 of Prioritizing Students with Disabilities
- 0:15and AI Policy, a resource that outlines a path towards AI systems that are inclusive,
- 0:19transparent, and designed with students, not around them.
- 0:23Erin Mote, CEO and founder of InnovateEDU, is a leading enterprise architect
- 0:28and systems change strategist dedicated to transforming the American education landscape.
- 0:33It's so great to have you on the podcast today, Erin. Thank you so much for joining us.
- 0:37Well, hi to my fellow Brooklynite, Aurora. That's right. Excited to join today's
- 0:44podcast, and thanks for having me.
- 0:46In December of 2025, EALA and New America brought together a cross-sector group to
- 0:51update our joint policy paper.
- 0:53That was such a fantastic convening. It's such a great event.
- 0:56And I just remember the excitement and the energy in the room that day.
- 0:59But tell us a little bit more.
- 1:01What did it feel like to be in the room where it happened? And why was it so
- 1:05critical to have so many different voices at the table for this topic?
- 1:09Well, I think I really agree with you that the energy in that room that day
- 1:14felt contagious. That's like the word that I would use, that folks we're really
- 1:18building on and upon each other.
- 1:21And part of it is, I think, the idea that we need to create space for people
- 1:26who don't often have the chance to sit around the table and who have a shared passion and commitment,
- 1:33and ethos around serving all students, particularly students with disabilities,
- 1:39to wrestle with some really hard topics and messy topics. And I think.
- 1:45The issue of AI and education is messy in itself.
- 1:49The issue of AI and education and students with disabilities feels like a challenge
- 1:57in managing both the promise and the peril of this technology.
- 2:01Because for students who are marginalized, so students who are traditionally
- 2:07not well served in our educational institutions, and unfortunately,
- 2:11that's many students with disabilities, the risk,
- 2:15can be really high when a new arrival technology enters the space.
- 2:20And the potential as well for the technology
- 2:25to have a positive impact on connecting data, new resources and opportunities,
- 2:32opening lanes of accessibility and inclusivity is also there.
- 2:36And so I think the energy in the room of wrestling down this promise and peril
- 2:42of AI, the simultaneous holding of that tension
- 2:46with folks who were coming from industry, from practice, from research,
- 2:52from the students with disabilities community, with parents and students and advocates.
- 2:58It was really, really, I think, a first time for many folks in that room to
- 3:04hear diverse perspectives.
- 3:06And to think through what could we do to bend the arc of this technology more
- 3:13for promise and less for peril.
- 3:15Yeah, I love that idea of bending it towards promise, less for peril.
- 3:20And like you said, like AI and education isn't new.
- 3:23This is not, this isn't a new challenge. But what were some of the specific
- 3:28like alarm bells or an urgent challenge that really set off the spark to create updated guidance?
- 3:35Well, I think AI isn't a rival technology. It's not an adoption technology,
- 3:40which means it's going to fundamentally reorder the way we work,
- 3:43we learn, we live, and we interact with each other. It's like electricity or the Internet.
- 3:48You know, I don't think 20 years ago we imagined that we could call a car to
- 3:54pick us up from our house from our phone. And the car would have no driver in it.
- 3:59And the car might not have a driver in it, right? Like, those are things that
- 4:04are changing the way that we interact in our daily life.
- 4:08And I think people feel that level of challenge with AI as an arrival technology.
- 4:15I think they're recognizing that like electricity or the internet,
- 4:19it is really going to fundamentally reshape our society.
- 4:25And I think with technologies like that before, right?
- 4:29Whether or not you have access to that technology is proximate to your wealth,
- 4:35your geography, and your economic opportunity.
- 4:38And so I think the idea that we have to design.
- 4:44For the sort of widest use case possible when we're thinking about these tools,
- 4:49when we have to think about in our design and our execution,
- 4:53how we protect the safety of young people's data.
- 4:58And in this case, students with disabilities, they're potentially their IEP
- 5:03data or confidential health data.
- 5:06How do we think about privacy by design, safety by design?
- 5:10And I think the arrival of AI in our classrooms now and the usage statistics in the high 80s.
- 5:18Really was a clarion call to what would happen if we had a conversation,
- 5:24not at the sort of next chapter of AI around how do we protect students with
- 5:31disabilities, but what if we made that a foundational design constraint?
- 5:36And I think that was both really exciting and really challenging,
- 5:40but I think, you know, you've heard me use this analogy before that we need
- 5:45to shovel the ramp, not just shovel the stairs, that when we shovel the ramp, everybody can get up.
- 5:50When we shovel the stairs, only some people can get up.
- 5:53So for people in Minnesota, Wisconsin, and Michigan, they really get that.
- 5:57For people in Arizona, I'm talking about when it snows and you got to like shovel
- 6:02the ramp in order to get to school, not shovel the stairs. And so these are
- 6:07things that I think we really can do.
- 6:11We can be purposeful in meeting this moment by the way we engage,
- 6:16design, and protect young people. Yeah.
- 6:19Yeah, that idea of safety by design, not as an add-on, not as an addition,
- 6:24not as an afterthought, and centering students with disabilities,
- 6:28not as an afterthought, but right as part of the design.
- 6:32So this is version two of the paper that we released in April.
- 6:36What are some of the big shifts or the big changes between V1 and V2? Yeah.
- 6:42Well, I think the first thing is that if people compare V1 and V2,
- 6:48they'll hear us talk about foundational pillars.
- 6:52And frankly, this idea that safety is table stakes for any of this work,
- 6:57no matter who we're talking about, whether that student has a disability or not.
- 7:02We have a fundamental duty of care to protect young people and educators' data, so adults too.
- 7:10I think the other thing that you would notice that's different is we were very
- 7:16focused on practice examples. One of the benefits that we have is we have some
- 7:21school districts who are taking different approaches around AI.
- 7:25Some are thinking about how do they prepare educators first and students next.
- 7:31Some are thinking about their role in mitigating foreseeable harms.
- 7:36I think the other thing that you would really notice about this work is that.
- 7:43Existing civil rights protections need to be at the core of the way we are considering
- 7:49design, development, and deployment of this technology,
- 7:53including things like wrestling with consequential decision-making,
- 7:59ensuring that if AI is engaged in any sort of decision framework,
- 8:03that there's a human in the loop.
- 8:05And these are things that I think we had the benefit of some of these use case
- 8:12examples from the field and those actual educators and parents and families being in the room
- 8:18saying, this is the promise of this technology.
- 8:21I want to support those incredibly positive use cases.
- 8:26And here's the things that I'm deeply worried about and fearful of.
- 8:30And I think, like, again, we just have a little bit more perspective around,
- 8:38how do we get clearer about, for us, what are just foundational table stakes,
- 8:43civil rights and safety, protection of young people's data.
- 8:47Yeah, that was a real helpful sort of like North Star when we were working on this paper.
- 8:53The way that you brought this up, this idea of this is civil rights,
- 8:57right? I mean, protecting student data privacy safety. And that's really what
- 9:01I feel like drove this iteration.
- 9:04You mentioned some of the case studies that we referenced and we worked with
- 9:08folks live in the room specifically.
- 9:11It's because we know it's easy to talk about challenges and the world,
- 9:14but there's a lot out there that is working.
- 9:17What are some promising use cases or examples in the field that you've seen?
- 9:21Yeah, I'm going to talk about two. One is like a tool and the other is an implementation framework.
- 9:27So the first one, I think, just like really resonates with me as a mom of a
- 9:32student with disabilities and a mom of a student who has dyslexia.
- 9:38You know, I think I've been very public with my son Robert's permission and
- 9:42talking about the struggle that we had in the beginning to
- 9:47get him screened and get him identified as a student with disabilities and to
- 9:52really hone in on the exact reading difficulties that he was having.
- 9:58And this is as somebody who has been an educator, built a school,
- 10:03like, is deeply familiar with this issue area, is an advocate for students with disabilities.
- 10:09And so, like, I deeply feel that struggle both for educators and for parents in that advocacy work.
- 10:17So one of the things I'm most excited about is to see some of the tools that
- 10:21are being used to support screening for teachers with dyslexia.
- 10:27So that, you know, an AI comes in and does universal screening for the class
- 10:32of dyslexia. This is work that's being done out of the Institute for Education
- 10:37Sciences and the National Science Foundation led by the University of Buffalo.
- 10:41And what's really exciting about this tooling is that, you know,
- 10:45we can achieve the policy directive of universal screening for all students for dyslexia.
- 10:50Usually that would take a school, a counselor, a paraprofessional,
- 10:55a set of teachers weeks to do.
- 10:58This AI tool can screen a class in about a half an hour, 45 minutes.
- 11:05It over-identifies students.
- 11:07Who might have reading difficulties. So it's actually catching a wider net, not just dyslexia.
- 11:13And then what the software and the pedagogy and practice do is set up a human
- 11:17to come back in and to do a more detailed screening.
- 11:20And so it's really... I was just going to say human in the loop,
- 11:22right? Human in the loop, really leaning on the professional practice and pedagogy
- 11:27of trained educators, but really helping do that universal screening so that
- 11:32we can intervene earlier.
- 11:34And I think we know what the science tells us is that early intervention,
- 11:38particularly with reading disabilities, makes an enormous difference in the trajectory,
- 11:45of a young person's ability to get to grade level content, to chat,
- 11:52not just literacy, but full fluency.
- 11:54And so I'm really excited about that productivity tool that is student facing,
- 12:00but is really about like solving, I think, a teacher productivity,
- 12:05a school productivity challenge that will benefit all kids.
- 12:09The second one is a practice and i want to like call out our friends at metro
- 12:14nashville um you know i think in a time when we have this like really hard diacotomy between like.
- 12:23Doomers and accelerationists so like either you're anti-ai or you're poor ai
- 12:28you know i think that's an interesting phrase yeah i mean it's true yeah it's
- 12:33true like we We have this really false dichotomy of either it's like gonna, you know,
- 12:39kill us all or it's going to be the greatest thing that happens to humankind.
- 12:43And I think that is a lot of what people are hearing in the popular media.
- 12:47And, you know, I'm firmly in the position of scout right in the middle there.
- 12:51I think there is incredible promise and I think there is incredible peril.
- 12:54I like that. And I think Metro Nashville is, you know, as a district,
- 12:58I think, has been in the scout position where they're not, you know...
- 13:03Kind of pressing the gas and saying AI for everybody all the time.
- 13:08What they're saying is we're going to give our teachers professional development.
- 13:11We're going to build AI literacy with our educators. We're going to build their fluency with tools.
- 13:17And then we can have a conversation about student-facing generative AI tools in particular.
- 13:24And so I love that go fast or sorry, go slow to go fast.
- 13:29And more importantly, I think that like intentionality of not feeling like a pressure to adopt.
- 13:38The shiny new object or tool, but really to ask the foundational question,
- 13:43what are the learning experiences we want to create for our young people and our educators?
- 13:49Should technology be used or not?
- 13:52Should AI be used or not? And I think that deliberate practice of interrogating
- 13:58not the tech, but the learning experience that you want to cultivate for young
- 14:03people is really important.
- 14:05And then the last thing I'll say about Metro Nashville, which I think is also
- 14:08really important, is they're also asking the question about whether or not,
- 14:13technology would give access to a learning opportunity that a student otherwise
- 14:18or an educator would otherwise not have.
- 14:20And I think that is a courageous conversation to say.
- 14:25Looking at our traditional curricula, looking at what everybody has access to,
- 14:31is there a place that technology doesn't just enhance the learning experience,
- 14:35but fundamentally gets that access opportunity for all students?
- 14:40And I have a lot of admiration for that team and for also just like them being
- 14:45really courageous in sharing their approach, because I think there's so much temptation,
- 14:52to be seen as the like most innovative district or the most AI infused learning
- 14:59environment. Highest numbers of.
- 15:02Yeah. Yeah, right, right. There's like this temptation for that to be the thing.
- 15:07And I think that the decision they've made there is what's the best learning
- 15:13decision for our community, for not just our students, but for our educators.
- 15:18And that is a brave set of things to consider.
- 15:21And those are some of the best learning opportunities. I remember on one of
- 15:24the earliest EALA school study tours. And Erin, I think you were there in San
- 15:28Diego. We were at a school. Yeah. We were at a school. We had some great weather.
- 15:31Shot up to San Diego for that weather during that tour. That was great. The torrential rain?
- 15:38The torrential rain. i'm being sarcastic no no right right i sold me a bill
- 15:41of goods on that uh i know i was like did not deliver yeah,
- 15:46but we visited a school that was emerging right and they were really really
- 15:50honest about what they were what they were challenged with how they were working
- 15:53through it and to to to amplify what you said about,
- 15:57the it's really brave to to pull back the curtain and let your colleagues and
- 16:02let the let the community let the industry in to see what you're doing because it's...
- 16:08You got to see, you know, how it gets done, how the sausage gets made.
- 16:10Right. It's not always pretty.
- 16:12But the thing is, though, there's something so incredible about seeing it happen.
- 16:15And that's where the best learning
- 16:17experiences come from. So shout out to shout out to Metro Nashville.
- 16:20Shout out to Metro Nashville. And to your point, like, here's the thing we know
- 16:23about the science of learning and development and actually learning.
- 16:26And my mama used to say that learning is painful. And what she was describing
- 16:29is like, you need friction in learning.
- 16:33And friction is sometimes like what we call failure. Or friction is sometimes
- 16:37what we call like not running out the gate, that there's like tension in the work that you're doing.
- 16:44And I think that is the thing that's really important for us to remember that
- 16:49like if it's not a little bit painful as a sector, we're probably not learning. Yeah.
- 16:57Well, I'm going to bring us to our last couple of questions here.
- 17:01And this is a big one. So give me kind of your big takeaway here.
- 17:06How would you like to see this policy paper being used or influencing future
- 17:13ed tech development, implementation?
- 17:15How would you like to see this paper showing up, you know, in a year, down the line?
- 17:21Yeah. Well, I think we're going to have to rev it again.
- 17:24So that's the first thing I'll say. V1, V2, there's going to be a V3.
- 17:31I think for developers of technology tools and somebody who has built and designed
- 17:36and architected technology tools, I feel like I'm speaking to my brethren.
- 17:41You know, it's really easy to design for the average.
- 17:45But let me challenge you to, on your persona board, add a student who has to
- 17:51access content a different way, has to have a different set of,
- 17:56you know, ways that they are encountering a visual cue or an auditory cue.
- 18:02Or are you even enabling those accessibility features at the beginning of your
- 18:07AI design? And what do your AI personas even look like?
- 18:12You know, is it someone who looks like an average kid?
- 18:16Or are you really considering the fact that, you know, one in five kids in this
- 18:21country is a student with a learning disability? And so if you have five personas
- 18:25on your board, you better have somebody who has a learning disability.
- 18:29The other thing I'll say to our school districts and to,
- 18:35our states, you know, there's a lot of conversation right now about restricting
- 18:39technology in classrooms and restricting technology in schools.
- 18:45I'm glad you brought that up.
- 18:46Yeah. Yeah. I really, I appreciate the conversation that we're in about,
- 18:53is the tech we have in our classrooms effective?
- 18:58Is it meeting the needs? Is it accessible? Is it interoperable?
- 19:01Is it secure? And is it private? And I certainly understand and parents and
- 19:07stakeholders, and even students asking those questions.
- 19:10The thing that I think people need to consider is that,
- 19:14in our space, we have done incredible work,
- 19:19through technology to give students with disabilities access and opportunity
- 19:25to learning experiences that happen in general education classrooms.
- 19:30And what we know from learning is that kids being in general education classrooms
- 19:38is where the majority of students with disabilities are, and it's the most effective
- 19:42in really thinking through the.
- 19:45You know, getting kids on grade-level content. And so what I would just say
- 19:49is, as we're navigating this tech lash, just like I ask developers to think
- 19:55about students with disabilities, I'm asking policymakers to think about students with disabilities.
- 20:00When we're talking about broadbands, let's not make, let's make sure we're not
- 20:05taking away the very opportunity for access, the very opportunity that kids
- 20:11have to engage and have those learning experiences, because many of those experiences
- 20:17are moderated through technology.
- 20:20Including for my own son. Yeah. Yeah. I mean, I've heard you say,
- 20:24you know, kind of these widespread bans is a blunt instrument.
- 20:28I've heard you use that phrase. And I love that. That's because it's coming
- 20:32from a place of let's investigate. Is this the right use of technology?
- 20:36But yeah, there's a lot of nuance there.
- 20:40Yeah. And I think people are scared, Aurora. I think like they see AI and I
- 20:44think generally the American public is more fearful than they are optimistic
- 20:49about AI. And there's lots of reasons for that in our society.
- 20:55And we can, you know, we could spend a whole other hour talking about what's happening there.
- 21:00But I think when we react by banning something, it often doesn't generate the outcome that we want.
- 21:11And so, you know, I was just, I think you know this, in Australia and they're
- 21:16talking with lots of folks in government about their social media ban and some
- 21:21of the things they're seeing.
- 21:22And one thing that I would name is that they're seeing child predators really
- 21:27engage with kids more on social media and kids not reporting those instances.
- 21:33To their parents because of the
- 21:35widespread narrative that you're not even supposed to be on social media.
- 21:40So if you know any kid, yeah, you know that like when you tell them not to do
- 21:45something and they still do it, they're not going to run to you and say,
- 21:50hey, mom, I did the thing you told me not to do and this thing happened.
- 21:54And now I need help. And so I think we really need to think about the unintended
- 21:59consequences of broad scale bans that don't consider the nuance of what we hope for from technology.
- 22:07And so I really want us to move the conversation from screen time to screen value. Ooh.
- 22:13Ooh, I love that. Yeah, that's it.
- 22:16Before we go, how can our listeners stay connected with you,
- 22:21learn more about the work that we're
- 22:23doing? Our listeners know to go to educatingallwearners.org, of course.
- 22:27We're going to link the paper. But how can folks stay in touch with Erin Mote or stay up to date?
- 22:33Where in the world is Carmen Sandiego or Erin? There's always something new.
- 22:38I post a lot on my LinkedIn, so feel free to follow my LinkedIn.
- 22:41That's both, like, stuff you'll hear that I'm thinking about and doing as well
- 22:45as, like, when I eat too much licorice and really regret it.
- 22:48All of that all the time. The human element.
- 22:52The human aspect of me. You know, the thing is, like, I think not just follow
- 22:58Educating All Learners.
- 22:59There's some really incredible projects at Innovate EDU, like the EdSafe AI
- 23:03Alliance, if folks want to dig deeper into practical and tactical resources
- 23:08that districts can use around AI and education.
- 23:11It's a great resource and a great page, and maybe we can link it in the show notes.
- 23:16And the thing that I would also call out is the AIPolicyLab.org website,
- 23:21where some of those districts,
- 23:23who were at that convening, but many others who are in our national network
- 23:28of 17 state policy labs and now 40 district policy labs are publishing their
- 23:33policies, their toolkits, their...
- 23:37Handbooks, they're playbooks. Like, there's just such a wealth there.
- 23:41And so if you are in a district or you're out of state and you're really thinking
- 23:46about AI and education, it's great to have policy papers.
- 23:50We want you to read the policy papers. And then you're going to have to think
- 23:53about how do I digest that into something my teachers and others can understand.
- 23:58And AIPolicyLabs.org is a great place to see what actual districts are doing
- 24:03with actual tools and actual materials.
- 24:07Thank you so much, Erin. It's always a pleasure when our paths cross.
- 24:10And thank you so much for the time today and for joining us on the podcast.
- 24:14Thanks so much. Have a great day. That's it for us today, folks.
- 24:17Thank you to Erin Mote. And to learn more, visit InnovateEDUNYC.org.
- 24:22If you enjoyed this episode, don't forget to subscribe on your favorite podcast
- 24:26platform and follow us on social media.
- 24:28I'm Aurora from EALA, and this was a deep dive into prioritizing students with
- 24:32disabilities and AI policy.