How Should Schools Teach AI Responsibly?

A podcast for educators on Microsoft’s Thailand AI-in-education push, exploring teacher time savings, student engagement, responsible AI use, and what large-scale skills programmes really mean for schools.

Transcript

SPEAKER_00 0:00

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 a Microsoft Source Asia article called From One Classroom to a Nationwide Movement, Advancing AI Skills in Education. It focuses on schools in Thailand and Microsoft's broader effort to build AI capability through programs, partnerships, and teacher development, arguing that the story is not just about tools and classrooms, but about a much wider movement to train educators at scale and reshape what AI literacy might actually look like in practice. And this is one of those pieces where it helps to read it with two thoughts in your head at once. The first is that this is clearly a corporate success story. Microsoft is telling a positive story about its own programs, its own tools, its own partnerships, so we should be careful not to read it as neutral research. But the second thought is that even inside that there is still something important here for educators. Because when a company starts talking not just about product adoption, but about training 150,000 educators, teacher workload, responsible AI use, and the changing role of the teacher, that tells you something about where the conversation is moving. According to the article, this transformation is already visible in specific schools, including Watana Witaya Academy, where Microsoft's AI tools are described as helping students focus, read with greater ease and confidence, and become more engaged in the classroom. The piece also says teachers are finding that AI tools supporting different learning needs can reduce time spent on routine tasks and free them to focus more on student growth and progress. Now that is a very familiar promise. Less admin, more attention for learners, more tailored support, higher engagement. We have heard versions of it many times before in EdTech. So the question is not whether the promise sounds good, of course it does. The better question is what kind of educational shift sits underneath that promise? And I think the answer here is about teacher capacity. Because one of the biggest misconceptions about AI in education is that its main value is flashy output, lesson plans in seconds, worksheet generators, image tools, automated summaries, all of that is real enough. But the deeper value when it works well is not the output itself, it is the reclaiming of teacher attention. AI is helping us hold the complexity, so we have capacity for creativity. That is the more interesting story. According to the article, Microsoft is working through public-private partnerships with Thai government agencies, including the Office of the Basic Education Commission, the Office of the Vocational Education Commission, the Institute for the Promotion of Teaching Science and Technology, and the Electronic Transactions Development Agency, with a goal of training and certifying more than 150,000 educators across primary, secondary, and vocational levels in a single year. That is huge. And whatever you think of Microsoft's motives, that scale matters because the AI conversation in education has often been trapped at the level of enthusiastic individuals. One teacher here, one pilot school there, one conference talk full of promise, but the hard part is always system capacity. Can this move beyond the early adopters? Can it reach the middle 80%? Can it become part of normal professional learning rather than a side hobby for the digitally confident? That is what makes this article interesting. It is not really about one clever classroom use. It is about infrastructure for adoption. Now I would still want to know much more. What does certification actually involve? How deep is the training? How sustained is the support? Are teachers being helped to think critically or mainly being trained to use Microsoft tools? Those are important questions. But even asking them takes us to the right place. Because schools do not just need access to AI, they need structures that help teachers use it well, reflect on it and challenge it. The article includes some survey figures from Microsoft Thailand's AI for Teachers program. It says that among 1,414 educators surveyed nationwide, 67.2% were already using AI in the classroom. On average they reported save in four hours per week, and the article frames that is equivalent to 95 full-time teachers annually. It also says 76.3% reported noticeable gains in student engagement, while 88.6% of schools had implemented or were developing AI guidelines. Now again, this is self-reported survey data in a Microsoft program, so we should not treat it as definitive proof. But it is still revealing. Four hours a week. Just imagine that for a moment. Not as a statistic, but as lived school time. Four hours is the difference between constantly catching up and having room to think. It is the difference between rushing feedback and planning a better conversation. It is the difference between surviving and redesigning. And here is where school leaders need to be careful. Time saved is only valuable if the system lets teachers actually reinvest it well. If those four hours simply disappear into more admin, more compliance, more cover or more meetings, then AI has not transformed anything meaningful. It has just fed the machine. But if that time is protected and redirected towards student support, curriculum thinking, feedback intervention, or professional learning, then something genuinely important starts to happen. So the leadership question is not only can AI save time, it is what will we let that time become. The article also says more schools are paying attention to responsible AI use, including safety, digital resilience, and prompt design. And it describes teachers helping students use AI to brainstorm and summarize more effectively while also reporting better exam performance and work quality. Now this is where the piece becomes more educationally interesting, because the role of the teacher according to the article is shifting from simply teaching content to showing students how to formulate sensible questions, evaluate information from multiple sources, and verify the reliability of AI generated content. A teacher quoted in the piece says students must learn to think, analyze, ask questions, and use technology ethically and responsibly, while teachers themselves must keep learning too. That to me is the real heart of the story. Not AI as shortcut, AI as context for better questions. Because this is what AI literacy actually is. Not memorizing tool features, not become dazzled by productivity tricks, it is collaborative reasoning ability, knowing how to work with the tool without surrendering your judgment to it. Human in the loop, outputs as drafts not answers. Enhancement not replacement. And I think schools need to get much sharper about that distinction. There is a weak version of AI integration where students are simply allowed to use tools and the adults call that innovation. Then there is a stronger version where students are explicitly taught how to question, refine, verify, compare, and evaluate. The first creates dependence dressed up as modernity. The second can genuinely build capability. The classroom examples in the article are also quite telling. At Watana Wataya Academy and Damajarane Wataya School, the piece says traditional text-heavy materials are giving way to more visual and interactive learning with images, animations, games, mood checks, and quick quizzes. It also describes a cover dance class where students use AI to explore interpretation, costume design, choreography, and a final showcase while in tech classes students are learning Python, with one student reportedly winning a national competition. Now I actually like that this is not presented only through core academic subjects, because one of the dangers in AI in education discourse is that it can become very narrow, very essay centric, very obsessed with writing tasks and cheating. But schools are far richer than that. Creativity matters, performance matters, design matters, technical making matters, AI can play a role in all of those spaces too. The question is whether it deepens the process or just decorates the output. And this is where I would want schools listening to this to be really concrete. Do not just ask where can we use AI. Ask where does it amplify thinking? Where does it stretch creativity? Where does it make room for stronger feedback or more ambitious work? And equally where does it tempt students to skip the struggle that builds understanding? Because that line is everything. Outsource the doing, not the thinking. The closing message in the article is about AI skills as a bridge to future success, with a school leader arguing that when children are equipped for the future, they can become part of a strong workforce, develop problem solving ability, and become more self-reliant. And that is where I think we need just a little caution. Education cannot collapse into workforce preparation alone. Schools are about more than employability. They are about identity, judgment, care, citizenship, meaning. But even with that caveat, there is a useful message here. The world students are moving into will absolutely require new habits of mind around AI. The real question is whether schools treat that as an optional extra or as part of what it now means to educate well. So my takeaway from this piece is not that Microsoft has solved AI in education. Far from it. It is that large-scale AI adoption in schools will rise or fall not on product features alone, but on teacher development, thoughtful leadership, responsible use, and the willingness to redefine the teacher's role around questioning, evaluation, and human judgment. That is the part worth holding on to. Because the future of AI in schools will not be decided by the slickness of the tool. It will be decided by whether we help educators and students become better thinkers in its presence. That's all for today. Thanks for listening.