Here’s what teachers in Canada have said about their experiences with AI in the classroom

Since ChatGPT and other major language models burst into the public consciousness, school boards are developing policies, universities are holding symposia, and tech companies are tirelessly promoting their latest AI-powered learning tools.

In the race to modernize education, artificial intelligence (AI) has become the new darling of political innovation. While AI promises efficiency and personalization, it also introduces complexity, ethical dilemmas and new requirements.

Teachers, who are at the heart of learning alongside students, observe this transformation with growing unease. For example, according to the Alberta Teachers’ Association, 80-90% of educators surveyed expressed concern about the potential negative effects of AI on education.

To understand overall policy needs, we must first understand classrooms and current teacher realities.

As a researcher with expertise in technology-enhanced teaching and learning at the intersections of assessment, leadership and policy, I interviewed teachers from across Canada, along with Bachelor of Education student Erik Sveinson. We asked them about their experiences with generative AI (GenAI) in the classroom.

Their stories help contextualize the reality of AI in a K-12 context and offer perspectives on harnessing AI’s potential without undermining education as a human-centered enterprise.

AI policy and educational wisdom

This qualitative study involved 10 teachers (grades 5 to 12) from Alberta, Saskatchewan, Ontario and British Columbia.

We recruited participants through professional learning networks, teacher associations, and district contacts, seeking to ensure a variety of perspectives from different grade levels, subjects, and geographic locations.

We thematically coded the interview data and then cross-referenced it with information from a review of existing research on the use of GenAI in K-12 classrooms. We highlighted convergences or tensions between theories on assessment, approaches to teaching in technology-enhanced environments, student learning, and teaching practices.

During interviews, teachers described a growing gap between policy expectations and the emotional realities of classroom practice.

What we heard

The following themes emerged from our interviews:

• The assessment crisis: Long-standing assessment tools, such as the essay or take-home project, have suddenly become vulnerable. Teachers spend countless hours questioning the authenticity of student work.

  All teachers interviewed consistently reported that they had difficulty with their current assessment practices and how students can use GenAI in their work. Confidence in the reliability of the assessments has been a challenge. The majority of teachers said they felt they needed to consider students cheating more than ever, given advances in GenAI technology.

• Equity Dilemmas: Teachers are on the front lines of determining which students have unfettered access to the latest AI tools at home and which do not.

• Educators see both opportunities and challenges related to AI. Good teaching is about fostering critical thinking and human connections. Ninety percent of teachers surveyed faced complex challenges related to equity and how to best support critical thinking in the classroom while developing foundational knowledge. In particular, middle and high school teachers in core subjects reported that students used GenAI tools in their free time outside of class, without ethical guidance.

“Another thing piled up”

A teacher from central Alberta said, “AI is definitely helpful for my workflow, but right now it feels like it’s adding to an already impossible workload. The policy says: “Embrace innovation”, but where is the advice and support?

Classrooms are dynamic ecosystems shaped by emotion, relationships, and unpredictability. Teachers navigate trauma, neurodiversity, language barriers, and social inequalities while delivering curriculum and meeting student expectations for success.

Educators say there is little recognition of the cognitive load they already carry, nor the time it takes to ethically review, adapt and deploy AI tools. They argue that AI policies often treat educators as passive implementers of the technology, rather than active agents of learning.

A high school teacher from Eastern Canada said: “AI doesn’t understand the emotional labor of teaching. She can’t see the trauma behind a student’s breakdown.

This perspective highlights a broader observation: teachers are not resisting AI per se; they resist implementation that does not take into account their emotional expertise and contextual judgment. They want professional training initiatives that honor the human and relational dimensions of their work.

Burnout, professional erosion

This disconnect is not just theoretical, it is emotional. Teachers report burnout, anxiety and a feeling of professional erosion. A 2024 study found that 76.9% of Canadian educators felt emotionally exhausted and almost half had considered leaving the profession. The introduction of AI, without adequate training and support, only compounds this stress.

The Alberta Teachers’ Association is also increasingly concerned that if not implemented properly and with support for teachers new to the profession, AI could deskill the profession.

A Vancouver teacher said, “I am a seasoned teacher and I understand the fundamentals of teaching. For beginning teachers, when algorithms write report cards or generate lesson plans, what happens to teacher autonomy and the art of teaching?

Turning teaching into a checklist?

Overall, the interview responses suggest that what is missing from AI policy is a fundamental understanding of teaching as a human-centered profession. As policymakers race to integrate AI into digitalized classrooms, they are missing a critical truth: technology can’t fix what it doesn’t understand.

Without clear guardrails and professional learning based on the informed needs of teachers and students, AI risks becoming a tool of surveillance and standardization, rather than a tool of empowerment.

This tension between innovation and deprofessionalization appeared in many teacher responses. Teachers expressed optimism about AI’s potential to reduce workload, but also deep unease about how it could erode their professional judgment and relational role with students.

A teacher from Northern Ontario said: “There is hope with new technology, but I fear AI will turn teaching into a checklist. We are not technicians, we are mentors, guides and sometimes lifesavers.

Teachers fear that without educator-led frameworks, AI could shift teaching from a human-centered practice to a compliance-oriented practice.

Responsible AI Policy

If we want to harness the potential of AI without harming education as a human-centered, student- and teacher-centered enterprise, we need to rethink approaches to AI innovation in education. It starts with listening to teachers.

Teachers should be involved in the design, testing and evaluation of AI tools. Policies must prioritize ethics, transparency and fairness. This involves regulating how student data is used, ensuring that teachers can check for algorithmic bias and ethical implications and also protecting their discretion.

Third, we need to slow down. The pace of AI innovation is dizzying, but education is not a startup. It is a public good. Policies must be evidence-based and grounded in the lived experiences of those who teach.

This article is republished from The Conversation under a Creative Commons license. Read the original article.