Students as AI Creators, Not Just Consumers

AI use in the classroom is no longer a niche behavior, it’s the norm. According to K-12 Dive, the share of middle schoolers using AI for homework jumped from 30% to 46% between May and December 2025 alone. High schoolers climbed even faster, from 49% to 63% over the same seven months.

We can see this trend growing throughout 2026. But how are they using it? What is their relationship with AI and how are they learning, thinking and creating with AI? This blog breaks down, what AI is being used for by students, how schools can increase their control on AI, and the impact AI has towards STEM learning.

How many students get formal training on using AI?

Despite AI adoption across the nation, data from the national deck states only about 1 in 3 middle and high school students know that their school have policies in place regarding AI use for thinks like homework, and task completion, further more they have not had clarity if the AI generating informations the students are receiving is trustworthy or accurate. 

53% of students report being afraid to use AI in their learning for fear of being wrongly accused of academic dishonesty, despite mostly using it for legitimate purposes like concept explanation and research ideation. That anxiety itself is a signal that schools haven’t yet built clear, confident frameworks for what appropriate AI use actually looks like.

AI is a controversial subject, there are positives and negatives on the concept and how it impacts learning, as with every new piece of technology. 

However, there is evidence to say, AI is not a Fad. It’s a concept that will keep evolving and be a tool used by the future generation Beyond the Classroom. WIth out correct and safe implementation and lack of AI teaching strategy, there comes a training gap risk, not a gap in access to AI tools, which are now widely and often freely available, but a gap in the sophistication of how students are taught to use them, potentially, impacting their employability.

How Students are Using AI?

We have broken how students use AI into categories – Consumer and Creators.

What is an AI Consumer?

AI Consumer is someone who uses the technology as an answer engine — asking a question, receiving a polished response, and using this as a basis to move forward in their task. 

What is an AI creator?

An AI creator is different, we don’t mean, they create their own versions of ChatGPT or Claude. Instead, they build their own AI-powered projects, tools or Agents; understanding how the underlying models generate responses; questioning those responses, developing their own judgement, quality checking AI answers, becoming creative and critical thinking. 

UNESCO and the OECD have both emphasized this distinction in their most recent education guidance, noting explicitly that better task performance using an AI tool does not automatically mean deeper learning has occurred – the depth depends on how the tool is used, not just whether it’s used.

Why this matters specifically for STEM

STEM education has a natural and important role to play in shifting students from AI consumers to AI creators, because STEM disciplines are where the technical understanding of how AI actually works – data, algorithms, model training, bias – naturally lives. A student who understands basic computational thinking and data concepts is far better equipped to critically evaluate an AI system’s output than one who only knows how to type a prompt into a chat interface.

This is precisely the shift education leaders are now pointing to as essential workforce preparation: as AI reshapes the American workplace, with workers, managers and AI assistants increasingly collaborating to complete tasks, students who understand AI as a collaboration tool — something to direct and build with — will be far better positioned than those who only ever learned to use it as a shortcut.

How can schools teach students to create with AI?

1. Move from prohibition to structured guidance. 

As outright AI bans soften across K-12 and higher education, the more effective approach emerging is teaching students not just how to use AI but how to question it — understanding how it generates responses, where it’s likely to be wrong, and how to evaluate its output critically rather than accepting it at face value.

Build hands-on AI creation into STEM curriculum directly

 Rather than treating AI literacy as a separate digital-citizenship unit, integrating actual AI-building experience — training simple models, building basic AI-powered projects, understanding how training data shapes output — into existing STEM coursework gives students the creator-side experience that consumer-only AI use never provides.

Address the training gap explicitly, for teachers as much as students. Given that fewer than half of both teachers and students report receiving meaningful AI training, closing the sophistication gap with tools like VR requires investing in teacher capability first, since teachers can’t model creator-level AI use for students if they haven’t developed it themselves.

H3: Design assignments around process and judgment, not just output. Placing more emphasis on how a student used AI, what judgment calls they made, and how they verified or refined AI-generated content shifts the incentive structure away from pure output-extraction and toward the kind of active, critical engagement that defines creator-level AI use.

Where EduverseSTEM’s AI tools fit this shift

Eduverse+’s AI content generation tools give students and teachers direct, hands-on experience with AI as a creation tool rather than a black-box answer engine – building 3D models and interactive content through AI-assisted workflows that make the underlying process visible rather than hidden. This kind of applied, project-based AI experience, embedded directly within STEM content students are already engaging with, is exactly the kind of creator-oriented exposure that current research suggests is missing for the large majority of students who’ve adopted AI tools without ever receiving structured guidance on how to use them well.

As AI adoption in classrooms continues accelerating through 2026, the schools that treat this moment as an opportunity to build genuine AI creators – not just confident AI consumers – will be preparing students for a fundamentally different, and more durable, kind of readiness.

Explore how Eduverse+ AI content generation tools give students hands-on AI creation experience, or see the full EduverseSTEM content library.