The Edtech Insiders Generative AI Map
A market map and database featuring 60+ use cases for GenAI in education and 300+ GenAI powered education tools.
The Edtech Insiders Generative AI Map
By: Ben Kornell, Alex Sarlin, and Laurence Holt
In the two years since ChatGPT launched, the education world has been polarized between techno-optimists and tech-skeptics who paint dramatically different pictures of our educational future— on one side tend to be technologists who often don’t understand teaching and learning and on the other, education leaders who distrust new technology of almost any kind.
But in between these extremes there is a whole spectrum of nuanced positions and perspectives. As technology people who care deeply about teachers and learners, how might the Edtech community cut through the noise and make sense of the emerging landscape of AI learning products?
At Edtech Insiders, we had a clarifying moment when we read Laurence Holt and Jacob Klein’s Map of Generative AI in Education in June of 2023. We had recently released our own Edtech Insiders AI Tools in Education Database, and placing our database alongside the map, we had our first comprehensive and strategic view of emerging AI in education tools in the post-ChatGPT-era.
What made Laurence and Jacob’s perspective so clarifying is that it didn’t start with the tech tools and companies, but rather with the teaching and learning use cases. These use cases not only pre-date AI, but are perennial and historic needs in education- the need for quality instructional materials for the classroom and home, the need for educator training and support structures, the need for collaborative, active and motivational learning experiences. It was an important reminder that, despite the exciting breakthroughs around AI, teaching and learning are still what matters most!!
Since then, we’ve been collaborating with Laurence and Jacob to update and fine tune the map to better understand where we are and where we’re headed with AI in education.
Today, we’re excited to release the next iteration of this work: The Edtech Insiders Generative AI Map (K12).
The new iteration of the Generative AI Map has expanded from previous version, now consisting of:
The Market Map: An at-a-glance snapshot of the 6 categories and 50+ use cases for AI in education, alongside leading tools that support each use case (and some emerging or potential use cases with few to no supporting tools yet)
The Generative AI Use Case Database: A repository and deep dive into the 50+ use cases for AI in education, including examples, details and supporting tools
The AI Tool Company Directory: A repository and deep dive into over 300 AI companies and tools in education, including each Gen AI company’s website, Linkedin page, company founders, target audience, sector (K12, HE, etc.), platform (web, app) and traction (web traffic and app downloads).
To get a full tour of the work, we’d love for you to take a look around and explore for yourself. Our Beta version is far from perfect, which is why we need your input and feedback! We hope for this to be a living resource for everyone interested in the evolution of AI in education, so please let us know what you think to inform future iterations and updates.
Don’t see your company on there? Don’t see a company in your portfolio? Found an error? There is a ‘suggestion’ form on virtually every page of the site to submit additional tools or comments!
Before you dive in, here’s a bit about our building methodology and key insights we learned throughout this project.
How We Built the Generative AI Map 2.0
In preparing for this new Generative AI Map, the team at Edtech Insiders started with Laurence’s thinking and models from v1 and v1.1. As we began to dream about what a 2.0 version might look like, we started with you, the audience, as our guide.
What would be most valuable to you as you contemplate AI in the education landscape? What insights at the macro and micro levels would be most helpful to you? What information could we pull together that will help you make meaning in your education context?
While we contemplated these big picture questions, it became immediately clear that the map needed to lead with use cases once again. Drawing from our own knowledge of teaching and learning, alongside what we’re seeing in the field, we added, edited, and reorganized the original map, expanding to six use case categories:
Within these use case categories, we include 60 total use cases for AI in education. While many of these use cases are the most established uses we are seeing for AI in education now, we also included emergent use cases that are currently less developed, but just as promising. These emergent use cases, which we find particularly interesting to emphasize as we look towards the future of the field, include but are not limited to:
For each use case, you can view a description of the use case, an example of the use case in action, and a number of technology tools, small and large, that currently support this use case.
Since each use case contains a number of aligned AI tools, our team worked to research and select the companies which are currently included (and will continue to evolve as the landscape changes AND as we discover and include new tools). To source the companies for the site, we began by pulling from Laurence and Jacob’s original AI in Education Map and The Edtech Insiders AI Tools in Education Database. In addition, the database draws from over a year of research along with conversations with Edtech Insiders community members, investors, thought leaders, entrepreneurs, and access from existing company lists from groups like Reach Capital, Andreessen Horowitz, Philippa Hardman, and others.
For each company we considered from these existing databases, we investigated further to decide what to include in the updated Generative AI Map.
Was the tool currently available and actively being used? How many people were using the tool? Was the tool being used in education in particular, or more broadly as an AI tool with a loose education bent? Is it a tool that is used in K12?
In the end, we limited the Generative AI Map to only include tools that are currently available (not in Beta or coming soon), that meet our web traffic guidelines with different cutoffs depending on company size, that have education as a stated use case on their website, and that have a K-12 clear use (even if it’s not their core sector in all cases)
You can view our full methodology section for even more details on these processes and parameters.
As it stands, our Generative AI Map now includes 315 companies. If that sounds like a lot, please be assured that it is still only a fraction of the AI education tools that exist. There are new tools being launched every day, and we will be working to keep the list as updated and comprehensive as possible, along with the help of our community. On every page of the site, there is the opportunity to suggest additional use cases and companies, which we will be reviewing and updating several times a year.
Key Insights
As we worked through our process, many insights emerged. Some of the key insights that stood out to our team include:
1) Market Recognition of Different Student and Educator User Needs
The original conceptions of AI in education often blended educator and student use cases: the same tool would be able to create classroom instructional materials and personalized student study materials. Today, we’re seeing a much more deliberate separation between AI tools for educators and AI tools for students. The driver for this boils down to core needs: AI use-cases for learners should (again, should) introduce an appropriate amount of learning friction, so the tools have to be trained to challenge learners, push them into their zone of proximal development, and not just give away answers. Teacher tools, on the other hand, derive most of their value from task efficiency, and their prime value prop is getting to the end product as quickly as possible without any friction.
2) “One-Stop-Shop” Tools for Teachers and Learners
Putting edtech products into use case categories has become increasingly challenging, as many tools are rapidly becoming “suites” of tools for teachers, or “one-stop shops” for schools with tools for multiple constituencies. The age-old wisdom for edtech companies was to focus on a single pain point as a wedge to drive growth, and then slowly expand your product over time. For AI technology, which rapidly increases development time, even with a small team, the reverse may actually be true. For one-stop-shop products, part of the benefit is that every educator or every student can find something of value in it, which makes procurement and implementation much much easier. The danger, of course, is that you may wind up with tools that are a mile wide and an inch deep.
3) New Use Case: Teacher Professional Learning
Over the last year we have seen a growing number of AI tools that place the educator as the learner, and address the scale and timing challenges of teacher coaching. Whether it’s leveraging simulations as part of teacher training, nudging research backed instructional moves, or providing space for guided reflection, professional development for teachers could actually be one of the clearest value-add areas for AI. As multimodal data capture expands, we can also see the quality and caliber of professional development improving exponentially.
4) Unlocking an Omni-Modal Future
New audio, visual, and video capabilities are creating new ways to support different learning modalities and to produce more engaging resources. On the delivery side, multimodal content has the potential to revolutionize student engagement, as tools are increasingly making or adapting instructional content into audio, video, gaming, interactives and simulations.
On the recognition side, visual capture (especially the ability to recognize photographs) has also become an integrated feature across many mobile platforms. Technologies that just recently have been a huge barrier to entry, like the ability to recognize student handwriting and spoken speech (including dialects) are fast emerging, with a few tools leading the way. We think they will become ubiquitous, and have the potential to completely change how we think about assessment.
5) Saturation Points and Waves
In accordance with Laurence and Jacob’s vision, the use case database aims to highlight not only the use cases that we already see in the market, but emerging use cases that are only recently being explored and may not have many (or any) example companies or tools whatsoever.
For some use cases, there are large and ever increasing numbers of tools, representing a first wave of tools that align with core GenAI capabilities. On the student side, common use cases include homework helpers, study material generation, and AI-enhanced tutors. Another increasingly popular category is Language Practice Partners, which allow students to converse with intelligent agents in foreign languages.
33 of the top 100 Education apps in the iPhone store address this handful of AI use cases. These apps are almost evenly split between “edtech incumbents” like Duolingo, Quizlet, Kahoot!, Chegg, Brainly, Khan Academy, and Photomath (a rare AI incumbent), and “AI-native” apps like Gauth, Question.ai, Answer.ai, Solvely, Learna and Praktika.
On the educator side, the first wave of use cases has revolved around the efficient creation, adaptation or review of instructional materials (use cases like lesson plan generation, AI-enhanced grading and feedback, quiz and question creation or administrative support to save teachers time). Now we are starting to see an exciting new second wave of teacher AI tools focused on omnimodal and interactive content, professional development, support for particular pedagogical approaches (project-based learning, work-integrated learning, small group facilitation), and support for specific populations (neurodivergence, underserved students, or ELL/MLL students).
6) Feedback
In many areas, AI is a poor substitute for human capabilities, but feedback is one of the most promising areas for student engagement with AI. This is largely because real-time, instant, and specific feedback, although known to be pedagogically impactful, is sparse for most learners and fully dependent on teacher availability. AI is increasingly good at this task, especially when it comes to feedback on written work. We expect to see even more instant feedback tools, including more nuanced and holistic feedback, and more feedback on visual and auditory work.
7) Metacognition and Process Evaluation
For generations, educators have required students to show their work or to hand in outlines or drafts so that they can support student’s processes rather than evaluating the end product. An increasing number of AI tool companies, from giants like Google (see The Metacognition Revolution), to startups like Sizzle, Querium and Snorkl, see the potential to elicit and act on student thinking, responding to the learning process, not just the finished product.
8) Coherence
Stepping back from the map as a whole, one final takeaway has to be the sheer number of new AI tools and the challenge of coherence as educators and learners navigate between them and their core curriculum and assessments. Per our latest article on the Learning Layer, we see tremendous opportunity for AI infrastructure to either stitch together the patchwork quilt of apps AND for their ability to accelerate feature parity across platforms, but as of yet, it’s still the wild west for the latest in AI.
Ongoing Questions
While we are unabashedly techno-optimists at Edtech Insiders, the Generative AI Map also connects to deeper concerns about the impact of AI on children and on education in general. Some of our ongoing questions as we iterate on and update the map include:
1) Academic Integrity
The ability to cheat with general purpose AI tools has created an incredible dilemma for educators at all levels. ChatGPT can bang out an A- essay in seconds and the proliferation of math problem solvers can even simulate the “show your work” steps of most assignments. No cheating detector system works well (although the integrity arms race is on), and so accusing students of using AI is fraught with social and academic risk. How does teaching and learning need to change to adapt to the realities of cheating technology? What guardrails can we expect or require for AI tools in schools?
2) Variable Quality
AI generated content has revived a deeper conversation around what constitutes quality curriculum and learning materials. Is the bar Teachers-Pay-Teachers? Or is it something higher? And how might we actually RAISE the bar of quality with AI?
4) General AI Tools
While we focused on purpose-built educational tools, let’s be real: ChatGPT is the dominant AI in education tool, perhaps matched only by Google’s suite with Gemini embedded. What is the appropriate role of ChatGPT in learning contexts? How do we differentiate when education specific tools are required?
5) Anthropomorphism
Across many of the tools we reviewed, learners are encouraged to build a relationship with the tool as if it were a human, character, or creature. This is frankly super confusing for learners, who are already being conditioned to anthropomorphize technology in all other aspects of their lives. And yet this strategy IS more engaging. How do we balance this?
6) Fundamental LLM Issues: Bias? Data Privacy? Hallucinations?
As the underlying LLMs improve their outputs, we have seen instances of hallucinations go down, but there are still plenty of examples of prompting sequences that generate alarming AI responses. Now imagine AI in the hands of a middle schooler - what could go wrong?!? More concerning, though, are continued examples of bias around race, gender, and ethnicity. And we still see gaps in privacy protection as children enter data into surface apps that communicate with the underlayers of AI. How do we develop legal and operational frameworks that protect children across all AI-enabled platforms?
What Do You Think?
The insights, questions, and concerns that came up as we developed this project are actually the whole point of the Generative AI Map. We hope that this starts the right discussions for you and your colleagues! That being said, our Beta launch of the Generative AI Map is really only as good as its value to you. We would love to get your input and feedback ahead of our full launch in 2025.
Areas where we would love your guidance include:
Use-cases that we are missing
Tools that we should add or recategorize
Research and efficacy studies that we should include
Flagging any information that is inaccurate or no longer true
And of course, anything that you really liked and would hope to see more of moving forward
Thanks so much for being on the journey with all of us, we can’t wait to hear from you!
Edtech Insiders is Sponsored By:
Since 2017, Tuck Advisors has been a trusted name in Education M&A. Founded by serial entrepreneurs with over 25 years of experience founding, investing in, and selling companies, we believe you deserve M&A advisors who work as hard as you do. Are you a founder thinking of selling your company? Have you received any UFO’s (unsolicited flattering offers) and need help determining if they’re real or hoaxes? We can help! Reach out to us at confidential@tuckadvisors.com.
Cooley LLP is the go-to law firm for edtech innovators, from early childhood through workforce. Informed by decades of experience in the education vertical, Cooley created the first edtech practice to provide industry-informed, business-minded counsel to companies and organizations at all stages of the corporate lifecycle. Cooley provides a multidisciplinary approach to client needs, offering seamless collaboration across offices and practices.
To learn more about what Cooley can do for you, reach out to Naomi May.
 | A guest post by
|
I am guessing this is strictly an American focus Gen AI tools map.
I would love to see more solutions from around the world, especially in the African space where we operate.
Great job by the way!
Cool stuff, insiders. Am I missing an ability to access the website directly from your directory? I can filter on use cases but then I need to Google the company name?