Sheryl, a year 10 student lacking direction, gets introduced to ClaudeCode by their IT teacher in an AI prompting lesson. With no previous coding experience, Sheryl develops a simple app in under a term. Sheryl releases it through the app stores and starts to see multiple downloads, growing a fanbase and a love of app design at the same time.

Bill, a year 9er, hands in an epic essay on the fall of the Nazis in World War 2. It is filled with accurate, well-placed research that supports the student's arguments. It is exceptionally well structured and insightful. All too insightful for a student who barely scraped through last year with a C average.

Elon, a kid you saw as harmless, but a bit of a class clown, goes too far down an AI image generation rabbit hole. He creates and shares deepfake explicit images of his classmates with a friend who then spreads the videos via multiple WhatsApp groups. The images spread like the plague, and before you know it, Elon is way out of his depth.

While the names above have obviously been changed, these are real stories, happening every day in schools across the globe.

AI tools are no longer a fringe technology, in schools or anywhere else, for that matter. They are being used every day, multiple times a day, on various devices by students across the globe. AI tools can put students on a path to design or coding that they never thought possible. They can be used to 'cheat' on an essay, producing content that the student doesn't even understand, reducing their critical thinking skills, and teaching them nothing about the writing process. And they can create AI-generated content that can cause serious damage to their peers and get them into serious trouble with both the education system and the law. However, with the right structures in place, all of these tools can be used to support learning rather than sidestep it, build confidence rather than dependency, and open up new pathways for creativity, research, and problem-solving that many students may never have explored otherwise.

While blocking ChatGPT and other generative AI tools was a strategy some schools took at the beginning of the AI influx (with varying levels of success), it is now widely understood that students need AI literacy, not blanket bans. With this in mind, we have put together a list of the top generative AI tools students are using in 2026, and what schools need to know.

It’s not just ChatGPT anymore

If you work in a school and still think students' use of AI starts and stops with ChatGPT, I have some news for you.

Students are not using one AI tool. They are building little unofficial stacks. One to brainstorm, another to rewrite, another to summarize, another to make flashcards, another to help with code, and another to generate images and video footage.

They might use:

• ChatGPT to brainstrom

• Grammarly to tighten sentence structure

• Quillbot to rewrite essays

• Perplexity to pull together sources

• Claude to help them code

• Midjourney to create an image for a presentation

AI use in education is no longer occasional or experimental. It is now woven into how students approach homework, revision, research, creativity, and problem-solving. The challenge for schools is that formal guidance is still catching up. The OECD’s Digital Education Outlook 2026 notes that generative AI is often used beyond institutional control because it is intuitive, versatile, and widely accessible, while HEPI’s 2026 student survey found that many students had already explored AI tools at school without receiving formal training or guidance on how to use them.

Recent survey data shows just how varied teen AI use already is. Homework is the most common use case, but students are also using generative AI to brainstorm ideas, write documents, create images and videos, summarize information, seek personal advice, and even generate content for jokes or to tease others.

That matters because AI use in schools is no longer confined to a single task, subject, or familiar chatbot. Students are using generative AI across the learning process for productivity, creativity, entertainment, and, at times, things that push well beyond what schools would consider appropriate.

A quick comparison of the main AI tools students are using

Before getting into the details, here is a quick comparison of the main categories of AI tools students are using, along with some common examples.

AI writing tools students are using

The most common AI writing tools students are using in 2026 are ChatGPT, Gemini, Claude, Grammarly, and QuillBot.

This is where most student AI use starts.

Students use these tools for:

• Brainstorming essay ideas

• Creating outlines

• Rewriting awkward sentences

• checking grammar

• generating summaries

• improving tone and structure

• researching sources

For many students (and users in general), using these tools doesn't feel like cheating or compromising their academic integrity. Apart from a few obvious edge cases, no one is typing in 'Write my full essay as a year 11 student' and hitting submit. It's not blatant cheating that's the issue, nor is it the use of AI in general. The concern is that when AI starts doing too much of the heavy lifting in the learning process, it undercuts students' critical thinking, weakening their ability to work through ideas independently and quietly replacing the struggle that real learning often requires.

Used well, AI writing tools can support students in the writing process. A student staring at a blank page with a serious case of writer's block can get started with an AI outline. A student who struggles with sentence structure and narrative flow may use Grammarly to check for grammar and clarity. A student trying to get a grasp on the subject matter might use ChatGPT or Claude to break down ideas before they begin writing.

Used badly, the same tools can flatten the entire writing process. Students can begin to outsource planning, interpreting, structuring, questioning, and expressing ideas in their own words. We have explored the hidden risks of AI in schools in more detail elsewhere, but the short version is this: when AI begins to replace thought rather than support it, the learning process suffers.

AI Research tools students are using

Students are using research-focused AI tools like Perplexity, Elicit, NotebookLM, and Gemini to generate research ideas, summarize academic research, and find relevant sources faster (something we all wish we had back in the day, right?).

Students use these tools to:

• generate research ideas

• explore research topics

• summarize complex concepts and academic research

• analyze data

• provide answers quickly

• build background knowledge

• Find relevant sources

This is one of the most appealing and enabling segments of generative AI in education. Research is notoriously slow, messy, and frustrating, especially for high school, college, and full-time undergraduate students who are still developing strong research skills. AI tools can make this first step considerably easier, providing studetns with a quick overview, suggesting research topics, and helping to break down unfamiliar ideas into bite-sized pieces.

While these tool sare genuinly useful, they can also create a false sense of mastery. A student who reads three AI summaries may feel like they understand a topic far better than they actually do. A student who relies on AI to provide answers may never develop the habits of source evaluation, comparison, and critical reading that proper academic research is meant to foster.

In higher education institutions, this matters even more. The pressure to produce high-standard work that is properly researched and cited, in very short time periods, makes the efficiency of AI particularly appealing.

This is the performance paradox of AI in education. Students appear to perform better because generative AI produce clear, polished work quickl.,However,  take the AI tools away, and performance falls off a cliff. As of 2025, nearly 80% of Australian university students reported using AI in their research, and researchers warn that this is creating an illusion of competence: the output looks strong, but the learning behind it is thin.

AI coding tools students are using

Coding tools like Cluade Code, GitHub Copilot, ChatGPT, and Gemini are helping students build apps, debug code, and develop technical skills more quickly.

How high school students are using AI in coding is genuinely exciting.

Students are using these tools to:

• Build simple apps

• Debug code

• Write scripts

• Test ideas

• Prototype projects

• Solve technical problems

• Build skills they might not otherwise have attempted

For some students, these AI technologies open a door that simply was not open before. A student who never saw themselves as technical can suddenly create something real. They can build an app, automate a task, or solve a design problem that would have felt completely out of reach six months earlier.

Used well, generative AI coding tools can help students learn by doing. They can provide studetns with personalized support, explain why something is not working, and keep momentum going when frustration would otherwise stop progress. Teachers are using coding apps to integrate AI into practical skills-based learning, helping students test ideas, troubleshoot, and build working projects, rather than just generating polished outputs without being able to explain the underlying workings.

AI image generation tools that students are using

Students are using image generation tools like Midjourney, DALL·E, and Canva AI for creative projects, presentations, and visual experimentation.

Students use these tools to:

• Create presentation graphics

• Generate concept images

• Make artwork

• Experiment with design

• Support media projects

• Create memes and visual jokes

There is a lot of creative potential in these AI-powered tools. They can help students explore design ideas, map out workflows, visualize concepts, and produce graphics without requiring advanced design skills.

However, this is also one of the highest-risk categories, particularly for curious high school students. These same tools can be used to create fake screenshots, explicit images, and humiliating edits of their classmates and staff. The ethical implications of these uses of artificial intelligence are well documented, as are the safeguarding risks. What once required significant technical skills and intent can now be done in a few prompts and a moment of poor judgment.

Conversational AI tools students are using

Students are using conversational AI tools like Snapchat My AI, Character.AI, and Notion AI for organization, quick advice, and everyday interaction.

Students use these tools to:

• Ask questions

• Get quick advice

• Roleplay conversations

• Fill time out of curiosity or boredom

• Interact with something always available

• say things they might not say to friends or family

• Organize tasks in a way that feels private and low effort

For many students, the appeal is not just convenience. Research from Common Sense Media found that teens often use AI companions becuase they are curious about the technology, find them entertaining, want advice, or like the fact that the interaction is available 24/7 and non-judgmental. Many teens share things with these 'bots' that they wouldn't tell friends or family.

What can schools do to support students using AI?

For both K-12 schools and higher education institutions, the old options of ignoring AI or trying to ban it outright are no longer sustainable. What schools need now is a clearer view of how students use AI, which AI tools are actually in use, and where AI use supports learning versus replaces it. That is where AI monitoring in schools becomes important, not as a way to panic about every new tool, but as a way to understand AI usage, identify patterns, and respond with clearer guidance.

Build AI literacy, not just AI usage policies

If students are already using AI tools every day, telling them not to use them is not an effective strategy. Schools need to teach students how to use AI responsibly.

• Question AI-generated answers rather than accept them at face value

• Compare outputs against trusted and relevant sources

• Understand that large language models can sound confident while being completely wrong (hallucinations)

• Recognise bias, misinformation, and incomplete summaries

• Understand the ethical implications of deepfakes, non-consensual image generation, and other harmful uses of AI-generated content

AI literacy should not sit in a one-off assembly or a dusty policy document. It needs to show up in everyday teaching and student-teacher interactions.

Use AI as a support system, not a substitution

Some of the most useful AI tools are the ones that provide additional support when a student is stuck.

That might mean:

• AI-powered tutoring for maths, writing, or revision

• Speech-to-text or text-to-speech tools for accessibility

• Tools that simplify complex concepts without removing the need to think

• Personalised support that helps students move at their own pace

Used well, AI can help students learn, build confidence, and access the curriculum more equitably. Used badly, it becomes a shortcut that replaces independent learning rather than supporting it.

The question schools need to keep asking is simple: Is this tool helping the individual student think more clearly, or just helping them finish faster?

Keep some tasks deliberately AI-free

If every task can be completed with AI, schools lose visibility into what students can actually do on their own.

That is why some tasks should deliberately remain AI-free.

These might include:

• In-class writing

• Oral explanations

• Handwritten planning tasks

• Source evaluation under exam conditions

• Coding tasks where students must explain their logic step by step

This is not about punishing students for using AI. It is about ensuring schools can still assess critical thinking, subject knowledge, and genuine understanding without a chatbot doing the heavy lifting.

Shift assessment toward process, not just product

AI has made finished work a less reliable signal on its own.

A polished essay, neat summary, or working app may still matter, but schools need to look more closely at how the student got there.

That means designing tasks that ask students to:

• Explain their reasoning

• Show drafts or stages of thinking

• Justify why they chose certain sources

• Reflect on how they used AI, if they used it at all

• Apply the same skill in a different context without AI support

The more the assessment focuses only on the final product, the easier it becomes for AI to mask weak understanding.

Use visibility as a teaching tool

If schools can see which AI tools students are using, they are in a far better position to respond sensibly.

That visibility can help schools:

• Understand which generative AI tools are actually in use

• Identify patterns in AI usage across year levels or departments

• Spot where AI-generated content may be affecting academic integrity

• Support conversations about acceptable use

• Make more informed decisions about AI integration

This is also where prompt monitoring can become useful as a teaching tool, not just a policing tool. If students are using prompts poorly, educators can use this as an opportunity to teach better prompting, stronger questioning, and more thoughtful use of AI. In other words, prompt monitoring can support prompt engineering, AI literacy, and better learning practices all at once.

Protect privacy, consent, and access to real support

Schools also need to be clear-eyed about privacy, consent, and human interaction.

Students should understand:

• What data they are entering into AI tools

• Where that data may go

• Why some tools should not be used with personal or sensitive information

• How AI can affect human interaction, decision-making, and help-seeking

The technology matters, but so does the environment around it. Students still need real teachers, real feedback, and real conversations. AI can provide answers quickly. It cannot replace judgment, care, or relationships.

Final thoughts

The AI tools students are using in 2026 are not just changing how work gets done. They are changing the learning environment itself.

Some AI tools help students build skills, explore new interests, and access personalised support in genuinely powerful ways. Others make it far too easy to shortcut the writing process, weaken critical thinking, or create AI-generated content that causes real harm.

That is why schools need more than vague AI policies and assumptions about ChatGPT. They need a practical understanding of the AI tools students use, how they use them, and the points at which support becomes dependency, misuse, or risk.

The goal is not panic. It is not blanket bans. It is clearer visibility, stronger AI literacy, and a more realistic understanding of how students are already using these tools day-to-day.