AI in edtech: Market map, funding and investor signals

Editor’s Note

This article was adapted by EUVC from an original insight published by Synaptic.

Read the original article.

Generative AI is changing how educational content is created, personalised and consumed. The opportunity spans lesson planning, grading, adaptive coursework, tutoring and self-directed learning. The constraint is equally important: institutions must capture those benefits without weakening academic integrity, learner trust or teaching quality.

For investors, the market is broad but uneven. Synaptic’s April 2024 analysis separates AI-native products from established education companies adopting AI, then examines where products, usage and capital were concentrating.

Synaptic identifies two primary user groups: people who create or deliver learning experiences, and people who consume them. The same underlying models can serve both groups, but the workflows and purchasing dynamics differ.

For teachers, higher-education instructors and workplace learning teams, generative AI can support:

The practical value is not simply faster content generation. Automating repetitive preparation and assessment tasks can create more time for mentorship, discussion and other human-centred parts of teaching.

For learners, AI can generate interactive material, adjust difficulty and pacing, explain concepts in different ways and provide immediate feedback. Chatbot-based tutors can also support homework, revision, exam preparation, language learning and independent study.

Synaptic’s thesis is that this moves digital learning beyond a single sequence designed for every student. The product opportunity is a learning experience that responds to the learner’s performance, needs and pace.

The same systems that help learners can also produce assignments on their behalf. That tension has pushed schools and universities to reconsider assessment design, acceptable AI use and how student work should be evaluated.

AI-detection software does not provide a complete answer. Researchers have documented limitations in detecting AI-generated text, while educators have raised concerns about false positives. Cornell’s guidance similarly recommends a broader approach built around critical and ethical use of generative AI, including transparency, reliability, privacy and oversight.

For product teams, academic integrity is therefore not a separate policy issue. It affects trust, adoption and the design of learning and assessment workflows.

Synaptic divides the market into two broad categories.

The market also spans several customer groups: K–12 and higher-education teachers, workplace learning teams, independent course creators and learners of different ages. That variety creates room for specialist products rather than a single category winner.

Virtual tutors represented roughly 40% of the AI-native tools in Synaptic’s market map. Products were being built for general tutoring as well as narrower use cases such as homework help, spoken-language practice and exam preparation.

The underlying product loop is consistent: deliver learning material at an appropriate pace, analyse the learner’s responses and adjust the next explanation, exercise or prompt. The quality of that feedback loop matters more than the chatbot interface alone.

Funding did not follow company count. Synaptic found 2.5 times more AI-native tools than AI-adopter tools, yet AI-native companies accounted for only 6% of the total funding dollars in its dataset. Their combined funding was equivalent to about 7% of the capital raised by AI-adopter companies.

Established platforms had existing distribution, institutional relationships and evidence that they could scale. Those advantages helped them attract most of the capital.

Within the market, virtual tutors and course-content generators represented about 60% of the companies Synaptic mapped but attracted almost 90% of funding. The virtual-tutor category also included established, heavily funded businesses such as Course Hero, Paper and Brainly, which increased its share of investment dollars.

Synaptic identified GSV Ventures, Reach Capital and Owl Ventures as prominent investors in the period, each participating in five to eight deals in the dataset.

Synaptic cautions that connecting a foundation-model API to a new interface does not create strong differentiation on its own. If competitors can access similar models, durable advantage has to come from elsewhere.

The article points to proprietary, use-case-specific data as one potential source of advantage. Fine-tuning models on relevant datasets can improve how a product handles a specialised workflow. Established companies may have an advantage because years of student work, assessment content or learner interactions can provide data that a new entrant cannot quickly reproduce.

Distribution and workflow position matter too. Education software can become entrenched inside institutions, so products already used for teaching, assessment or learning management may be able to add AI features more efficiently than a standalone entrant can replace the full workflow.

For additional context on proprietary data, workflow position and how software companies can respond as AI changes their category, see this related board-level framework for evaluating AI resilience.

In its 2024 website-traffic review, Synaptic highlighted MagicSchool, Twee and Gizmo as products gaining traction. Their use cases covered teacher workflows, English-language lesson creation and personalised study through flashcards, quizzes, spaced repetition and active recall.

Synaptic also highlighted positive customer feedback for AI-native products Curipod and Genei. Curipod focused on interactive lessons and feedback on student responses, while Genei supported research workflows such as extracting information, planning a written structure and generating citations.

These examples were signals from the 2024 snapshot, not a current ranking. Their broader significance was the range of workflows attracting demand across both instructors and learners.

Synaptic expected personalised learning to remain the most important entry point for new products. It identified adaptive study materials, quizzes, flashcards and virtual learning environments as areas where products could become more nuanced.

Education and workplace-learning software is often deeply embedded in institutional workflows. Synaptic therefore expected established platforms such as Kahoot!, Course Hero and Gradescope to keep adding AI capabilities or acquire specialist products.

Synaptic counted more than ten transactions in AI edtech from 2020 to April 2024 and expected further consolidation. Examples included the acquisition of Kahoot! led by Goldman Sachs Asset Management, Research Solutions’ acquisition of scite and Docebo’s acquisition of Edugo.AI.

The logic was capability acquisition: established businesses could buy specialist technology or teams to extend their products and defend distribution.

Synaptic’s analysis suggests five practical diligence questions:

In Synaptic’s framework, an AI-native company is built around AI as a core part of its product. An AI adopter is an established education company adding AI features to an existing platform or workflow.

In Synaptic’s 2022–24 dataset, virtual tutors and course-content generators attracted almost 90% of funding in the space, while representing about 60% of the market it mapped.

Synaptic’s analysis points to proprietary data, product-specific fine-tuning, workflow integration, institutional distribution and user trust. Access to the same foundation models is unlikely to be enough on its own.

Synaptic is the corporate author of the original article. Synaptic is a data intelligence platform for private investors, supporting multiple stages of the investment process from deal sourcing to portfolio management. It is used by venture capital, private equity and hedge fund investors.

Learn more about Synaptic’s research and tools for private-market investors.

This article was adapted by EUVC from an original article published by Synaptic.

Read the original article.