← AI for Educators Daily with Dan Fitzpatrick
AI for Educators Daily with Dan Fitzpatrick
Student Perspectives AI: Only 44% Think AI Homework is Cheating
00:10:28
Loading summary
Transcript1 lines
- [00:00]
A
If this episode makes you think, please let us know in the comments and support us by subscribing and leaving a review. Thank you. Today we are exploring student perspectives AI Diving into a really insightful report that reveals something quite surprising. Only 4 in 10 teenagers actually think it's cheating to use AI to complete all of their homework. Now that's a pretty striking number, isn't it? The piece we're looking at is called Navigating AI in Education Pupils Perspectives on the Role of AI in the Classroom. Published by Oxford University Press, this report draws on some significant research, including a survey of 3,113 to 18 year olds and additional qualitative studies with over 700 pupils across the UK, all part of Oxford's Educational Research Forum. What they've done is truly valuable. They've listened directly to young people about their views on AI, building on their earlier Teaching the AI Native Generation report, Dr. Alexandra Tomescu, Head of Product AI at Oxford University Press, and Dr. Sarah Ratner, Principal Investigator at AI in Education, Oxford University, really emphasized that student voice must be at the heart of this conversation. And I couldn't agree more. Too often we educators are making assumptions when we should be asking the source. So let's unpack some of these student perspectives. AI the first thing that really jumps out from this report is just how nuanced and honestly conflicted young people are about what constitutes appropriate AI use. While only 44% think using an AI tool to complete all homework is cheating, or almost one in five students also think it's cheating to simply ask an AI tool for homework tips. That's a huge spectrum of understanding right there. It highlights a massive gray area for them. What the research clearly shows is that only 15% of students feel they've been given enough guidance on this. Think about that for a moment in your own school context. Are we providing that clarity? We're talking about a generation that's grown up with these tools and they're asking for help to navigate the ethical waters of AI cheating homework. This isn't a battle against them, it's an opportunity to teach them to outthink machines, not just outsmart them. And this leads into the second really interesting finding. Students are actually quite hesitant to use AI as a default for their schoolwork. The qualitative studies, which involved Judith Gray, Market, research Manager at Oxford University Press, revealed that when given a task and offered the choice to use a generative AI tool, almost 3/4, 72% to be precise, decided not to use it. Now, that really surprised some of the Teachers who participated in the study. One deputy head in London said they thought students would be completely in favor of AI, but the responses were much more nuanced. Why this hesitation? Students gave some incredibly thoughtful reasons. One year, Tenala pupil Philip from the Mossland School said it wouldn't be actually what I truly think it would be. A vague interpretation of what I've told a robot to expand on, and it might not even make sense. Others said they'd rather their work be my own because AI doesn't know me as well as myself. This is huge, isn't it? This isn't about fear. It's about a deeply ingrained understanding of the value of are. In their own voice, their own thinking, they are protecting that human element without us even asking. It reinforces one of my core philosophies. Machines can compute, they cannot wonder, they cannot care. These students inherently grasp that. They're also incredibly clear about the irreplaceable value of their teachers. A remarkable 73% of students pointed to skills their teacher has that AI can never replace. They highlighted things like being a real empathetic human, understanding when a student is struggling and explaining things in a different way until it clicks, and a personality that brings the subject to life. This is the human in the loop principle playing out in real student sentiment about preserving and protecting those human domains like wonder, care, judgment, relationship, imagination, and wisdom. This isn't just about teachers making lesson plans. It's about the deep human connection that drives learning. Now, the third key takeaway for educators here is what students do want from AI and how teachers can bridge that gap. Over three quarters of school children, 77%, want their teachers to use AI tools and lessons. They're looking for AI to make complex work, easier to understand, to help with exam practice questions, and even to plan fun lessons. They see AI as a way to free up teacher time so they can get more one to one support. This aligns perfectly with the enhancement, not replacement pillar they want AI to augment, not erase the educator. This is where the conversation shifts from can they use AI? To how do we teach them to use AI? Well, what students are actually asking for is better AI literacy. They want guidance on how to write prompts more effectively, how to ensure they're not cheating, and how to use AI for meaningful revision. As one student put it, asking deeper questions prompts is good as it makes me think for the questions to ask to get better answers and, and that in turn makes me understand more. That's exactly what we mean when we talk about teaching students not to outsmart machines, but to outthink them. This isn't just about memorizing tool features, it's about developing collaborative reasoning ability, managing AI conversations with precision, and understanding its limitations. It's the essence of teaching the AI native generation how to truly think with AI. If you're finding these insights valuable, please do consider following or subscribing to the podcast. It really helps us continue to share these important conversations about AI in education. So what does this all mean for school leaders and teachers? On a practical level, this is our fourth point. Daniel Williams, Associate Assistant headteacher at Bishop Vesey's Grammar School, provide some brilliant practical tips for tackling what he calls AI anxiety among teachers. His advice resonates so strongly with the outsource the do and not the think and philosophy. He suggests starting with low risk teacher tasks, things like drafting parent emails, rewriting instructions, or creating a first draft of a worksheet. This allows teachers to build confidence privately with without immediately wading into assessment integrity or pupil data issues. He also stresses giving AI very specific instructions, moving from a vague make me a quiz on forces to something much more precise detailing year group, subject topic, exam board style, ability level and format. This is the application of prompting frameworks like prepared or prep in a real world context. The more context you provide, the better the output, which then serves as a true first draft. This idea of a first draft is crucial. Daniel Williams emphasizes that anything AI produces should be treated as a draft, not a final answer. He rightly states that AI can sound confident while being wrong. The teacher remains the expert, bringing their subject knowledge, understanding of pupils, and awareness of common misconceptions. It protects the human in the loop. Chris Goodall, Data and Digital Innovation Director at Bourne Education Trust, points out that if students are using AI to take shortcuts, that's a task design problem. This is a profound statement. It takes the pressure off catching students and puts it squarely on our pedagogy. It connects directly to the three Ps assessment model, product, process and performance, and the idea of cognitive stretch. We need to design learning that cannot be faked because it demands depth, care and imagination. When we focus on the process of learning, including how students interact with AI rather than just the final product, we encourage that productive struggle. Dr. Penelope Wolf, Director of Impact and Learning Design at Oxford University Press, reminds us that any AI tools developed for education need learning design principles at their core and are grounded in strong pedagogy. This really hammers home the purpose over technology pillar. We don't adopt AI because it's new. We adopt it because it serves a clear educational goal. This report is a clear call for proactive leadership. Teachers are asking for more training, clear guidelines and model policy documents. This isn't resistance, it's responsible professional caution. As Daniel Williams puts it, these educators need time and space, just as our change leadership principles outline. Give them that and they become the best drivers of innovation. We need to build from their strengths, anchor AI to existing successful practices and create champions who can lead the way. Ultimately, these student perspectives on AI are incredibly reassuring. They show us that our students aren't just blindly adopting new tech. They're thinking, they're questioning and they're looking to us, their teachers, for guidance. They value the human connection and their own unique voice above all else. That's a powerful foundation to build upon for the future of AI in education. That's all for today. Thanks for listening.