AI in Edtech: Use Cases, Common Challenges, and How to Overcome Them

Around 65% of students and 71% of teachers claim AI tools to be essential aspects of students’ success in college and at work.

With its strong potential to enhance the quality of education for both learners and educators, the global AI in edtech market is growing at an impressive compound annual rate of 31.2% and is projected to reach $32.27 billion by 2034.

The transformative potential of edtech AI software is acknowledged by both edtech app providers and educational institutions, as 40% of schools and districts in the USA have already adopted AI policies. With its powerful capabilities for automation, analytics, and enhanced decision-making, AI enables more efficient, inclusive, and immersive learning experiences.

In this article, we will examine the most common applications of AI in education, key obstacles to adopting it in edtech, and strategies for overcoming them.

In 2025, around 92% of students are using AI in education in some form, up from 66% in 2024. These numbers illustrate the fact that both learners and educators recognize the transformative impact of AI on educational outcomes.

A significant surge occurred in 2023. In particular, a report from the UK Department for Education revealed that in April, only 17% of primary and secondary teachers in the UK were using generative AI to aid with their schoolwork, but by November, this figure had already risen to 42%.

This growth followed the release of a web-based prototype for OpenAI’s ChatGPT on November 30, 2022. In January 2023, the solution’s user base had already reached 100 million, which turned out to be a moment when AI truly became a public technology. OpenAI needed time to roll out full-scale releases and mobile versions, and learners and educators also needed time to understand how to enhance the educational process with AI.

What followed was the growing adoption of large language models (LLMs) and AI-powered features in educational software. Such projects differ in their size and complexity, ranging from the release of an education-specific version of ChatGPT across 23 campuses of California State University to the launch of 148 new AI-powered language courses in the Duolingo mobile app.

With its wide range of applications, it’s no surprise that AI has become the leading trend in the global education industry. This aligns with the surging demand for AI-powered apps, which has recently seen a 19-fold increase in downloads — from 6 million in January 2023 to 115 million by December 2024.

Consequently, AI has moved beyond a novelty to become an integral part of modern education. Both open-source AI tools and edtech apps powered by AI features transform how students learn, teachers educate, and academic institutions operate.

AI serves multiple purposes in educational technology. The image below highlights the main advantages of using AI in edtech, as reported by teachers in Carnegie Learning’s State of AI in Education 2025 report.

Ultimately, AI functionality in edtech applications can generally be divided into three main categories:

• Automating educational workflows
• Enhancing personalization
• Making education more engaging and accessible.

Let’s discuss the main AI use cases for the following categories in more detail.

According to a 2023 report from the UK Department for Education, participating teachers and middle leaders spent an average of two hours per day on administrative tasks, such as managing enrollments and conducting routine assessments. AI has proven to be an efficient tool for automating routine educational workflows, with educators reporting that they can save 5–10 work hours per week using AI tools.

Common use cases where AI is used for automation by educators and, in some cases, by learners include:

• Automated grading and assessment. AI algorithms can evaluate multiple-choice and short-form answers. Learning management systems, teaching assistance solutions, and online assessment platforms enhanced with such AI algorithms provide immediate feedback and reduce teacher workload. In particular, artificial intelligence can reduce grading time from six hours to 20 minutes, which allows educators to focus more on thoughtful lesson planning and meaningful student engagement.
• Schedule recommendation. Some learning management and planning platforms, as well as virtual tutors, have AI-powered features for planning schedules. These tools use artificial intelligence to optimize class timetables, allocate resources, and efficiently personalize learning schedules for students and educators.
• Learning analytics. AI can assist in learning analytics and teaching assistance platforms by efficiently organizing large volumes of data and retrieving insights on learners’ academic performance. Such AI tools can help teachers identify struggling students early and provide tips on keeping them motivated and engaged in the learning process.
• Automated transcription and note-taking. AI-powered recognition systems can generate detailed lecture notes and even transcripts prepared with voice recognition technologies. Such functionality can be included in various tools, such as learning management systems or virtual tutors. It helps students review content, stay focused during classes, and access accurate summaries of lessons they missed.

A possible option for automating educational workflows is integrating an edtech solution with an existing AI-powered tool. For instance, a learning management system or a virtual library can integrate AI-powered video transcription and summarization that will automatically convert video content, such as lectures, training sessions, and onboarding materials, into structured transcripts and AI-generated summaries. This allows educators to save time both in delivering key lecture points and in assisting students with catching up on missed lessons.

Learners who receive individualized educational experiences show significant improvements in motivation and academic success. With its advanced analytics and content-generation capabilities, artificial intelligence offers unprecedented opportunities to design truly tailored learning experiences.

These are specific use cases where AI features included in edtech software can enhance personalization:

• Personalized learning paths and skill mapping. Such functionality can be applied in a wide range of edtech solutions, including learning management systems, virtual tutors, teaching assistance platforms, and apps for learning languages. AI algorithms can assess students’ progress and adjust the content, difficulty, pace, and next steps in their educational journey. accordingly. This ensures that each learner stays challenged but not overwhelmed.
• Intelligent content generation. AI tools can generate dynamic, tailored study materials, such as quizzes, flashcards, and summaries, based on specific learning goals. Such functionality is especially relevant to learning management systems, teaching assistance solutions, virtual tutors, or gamified learning apps.
• Real-time personalized feedback and hints. AI-powered learning management platforms, language learning apps, or virtual tutors offer contextual feedback and step-by-step guidance during exercises. This feature helps students understand their mistakes immediately and adjust their learning patterns accordingly.

For example, Google’s virtual tutor, Socratic, uses advanced machine learning algorithms to provide personalized learning experiences. The platform analyzes user interactions to identify individual learning patterns and areas of difficulty. Socratic uses this information to adapt its responses to deliver a tailored approach that addresses each learner’s specific needs.

AI in education technology helps break down linguistic, sensory, and cognitive barriers to quality learning through adapting educational content to the needs of specific learner categories. AI tutors can also increase accessibility and affordability of education for learners who face challenges in finding or affording human tutors.

While AI-powered adtech solutions cannot fully replace the personalized guidance of human instructors, they can efficiently support individual learning by providing tailored assistance and helping users build foundational knowledge across various subjects and disciplines. In addition, by enabling learners and educators to transform educational content into more accessible formats and by enhancing simulations and gamification, artificial intelligence can make learning more engaging.

Here is a list of common ways AI is used to create a more accessible and engaging learning environment:

• Content captioning and summarization. AI can generate accurate captions for educational videos and create concise summaries of lengthy texts. This can make learning materials more accessible to the learners, as well as help them capture essential information more efficiently. AI functionality for content captioning and summarization can be applied in various categories of edtech apps, including learning management systems, virtual tutors, teacher assistance solutions, etc.
• Real-time translation and multilingual support. Such functionality is widely applied to various learning management systems, virtual tutors, and virtual libraries. For non-native speakers, AI enables instant translation of instructions, lectures, and learning materials. This makes global content locally accessible. Notably, AI-powered translation features often lie at the core of apps for language learning.
• Gamification and simulation. AI often forms the core functionality of gamified and simulation-based learning tools. Artificial intelligence algorithms enable immersive simulations that replicate real-world scenarios relevant to the field of study. For example, AI can automatically generate challenging road scenarios in an educational driving simulator, requiring learners to make real-time decisions. Coupled with gamification elements, such as points, levels, and challenges, these simulations increase engagement, motivation, and knowledge retention.
• Speech-to-text and text-to-speech. Virtual tutors, language management systems, or language learning apps often include AI algorithms that support learners with visual or auditory impairments. For instance, voice retrieval-augmented generation (RAG) technology can convert spoken language into written text. Meanwhile, computer vision and optical character recognition (OCR) can transform printed or on-screen text into natural-sounding speech. Such features may also be beneficial when students and educators, for varying reasons, need to transform educational materials into a different format.

To improve the accessibility of an edtech platform, it can be integrated with an existing AI-powered solution. For example, integrating an AI-powered video translation solution can make multilingual content in virtual libraries more accessible to a diverse range of learners. This enables educational institutions to localize course materials, lectures, and training videos for international students while preserving the instructors’ original tone and teaching style.

Despite growing interest in the potential of artificial intelligence for educational technologies, several obstacles continue to hinder its progress. For instance, 64% of educators in the UK said they don’t know enough about AI to apply it effectively in their roles. The image below highlights the key concerns educators have regarding the adoption of AI in education.

Most of these problems are associated with the use of open-source AI tools. Additional challenges arise specifically in the custom development of AI-powered edtech solutions.

According to the IBM AI adoption survey, 42% of business executives cite the lack of expertise as one of the biggest obstacles to implementing generative AI. Skilled data scientists, AI strategists, and AI or LLM development specialists are vital for schools, universities, or edtech providers aiming to build AI-powered tools that bring actual value. Finding specialists skilled in both AI and edtech software development can be a significant challenge.

A possible solution involves outsourcing development to a dedicated team of AI specialists with edtech experience. This allows you to fill out the possible expertise gap, save time and budget required for hiring and training an AI software development team in-house, and avoid common AI software development mistakes, such as lack of scalability planning.

The lack of available proprietary data is a significant problem in AI adoption, cited by 42% of business executives from the IBM survey. While the edtech domain generally has ample data, significant challenges remain regarding its quality and organization. Information from different academic sources is often inconsistent or unreliable, and incomplete auditing of learners’ performance can pose a major issue for AI-powered educational analytics.

To overcome this problem, it is vital to ensure that custom edtech software relies on accurately organized and labelled knowledge bases compiled and curated by specialists with deep industry expertise.

Around 35% of educators in the UK are concerned about the risks of adopting AI, particularly around data privacy and confidentiality. Edtech platforms often handle sensitive student information, such as personal data and behavioral analytics, which impose specific data protection obligations.

To handle this problem, educational institutions or AI edtech companies must ensure compliance with regulations like GDPR or FERPA. This involves implementing advanced data encryption, authentication, and monitoring practices for the entire system with AI features. Such an approach will help you safeguard the solution against vulnerabilities like data leakage, prompt injection, and adversarial attacks, which can threaten user data and software performance.

To proactively identify vulnerabilities in your existing edtech solution, conduct periodic technical software audits.

AI algorithms can provide unreliable outputs, which leads to a general decline in trust towards them. In particular, skepticism toward the information generated by AI has increased among millennials, from 21% in 2023 to 30% in 2024. In the educational setting, this can cause serious issues, as learners may receive inaccurate or false information and be exposed to potential ethical shortcomings of artificial intelligence.

To prevent misleading AI model outputs in edtech, it is essential to focus on data organization and curation so that your ML models will base their responses upon credible and easily findable data. Additional steps to mitigate AI bias in education include regular audits and continuous performance monitoring. These practices enable a proactive approach to AI hallucinations and model degradation.

AI revolutionizes the edtech industry by enhancing educational software with advanced automation capabilities, helping both educators and learners save time on routine workflows. In addition, artificial intelligence algorithms ensure learning personalization and can make educational content more engaging and accessible to learners with technologies like OCR and RAG.

However, particular challenges, such as a lack of expertise, issues with data organization and quality, security concerns, and AI bias, still create obstacles to a wide implementation of AI in edtech software. Strong domain expertise and the right approach are required to overcome these problems.

The Leobit team combines extensive expertise in implementing AI and LLM solutions with strong expertise in providing edtech app development services. For example, our portfolio includes several notable edtech development projects, such as custom development of an educational and consulting platform for people with disabilities and a discovery phase for an externship coordination platform.

As a Microsoft Solutions Partner and a company that has successfully applied several AI agents to support our internal processes, we possess a deep understanding of AI technologies, their promises, and ways to solve common AI adoption challenges.

Contact us to transform the educational process through advanced AI technologies.