The question we’ve been asking students for two years — “did you use AI to write that?” — just landed on the other side of the desk. It’s just as hard to answer.
For two years, the main AI conversation in schools has been about students. What they’re using, what they’re hiding, whether the essay in front of you is actually theirs. Teachers have been the ones setting the rules, designing the guardrails, flagging the suspiciously clean paragraph. The conversation has run almost entirely in one direction: adults trying to figure out what to do about kids and AI.
This week, Education Week published a piece that flips it. The question: if you use AI to help craft your professional assessment portfolio — the reflective documents teachers produce for licensing, professional development, and evaluation — is that cheating?
It’s a harder question than it looks — and one most teachers have been sitting with for longer than the news coverage suggests.
What a Portfolio Is Supposed to Do
Teacher assessment portfolios aren’t like student essays. They’re not a snapshot of what you know at a moment in time. They’re meant to capture professional growth — reflections on practice, analyses of student work, evidence of instructional decision-making. The portfolio is supposed to be the teacher thinking on paper about teaching.
Which is why the AI question lands differently here than it does with student work.
When a student uses AI to write an essay, the concern is that the demonstration of learning is hollow — the skill being assessed is exactly the one being outsourced. When a teacher uses AI to help draft a portfolio reflection, what’s being outsourced? Is it the thinking? Or is it the transcription of thinking that happened in real classrooms, with real students, that the teacher genuinely did?
That’s not a clean distinction. It’s a genuinely murky one. The EdWeek piece is honest about that — critics have long argued portfolios are vulnerable to outside help, from coaches who edit lesson videos to companies that review submissions for a fee. AI doesn’t create that problem. It makes it cheaper and more available.
But “more available” is exactly where it gets interesting.
The Standard That Runs Both Ways
Teachers have spent the last two years developing strong intuitions about what student AI use looks like. An essay that shows none of the writer’s characteristic voice, that produces clean argument structure without the messy thinking that usually precedes it, that arrives a day after the student was publicly struggling — that’s worth questioning.
But those same teachers are now in a position where AI tools can help them organize reflective writing, improve clarity in their portfolio narrative, and frame instructional decisions in the precise language that evaluators expect. Many of them are using those tools. The institutional accountability that follows student AI use — the policies, the conversations, the formal scrutiny — hasn’t extended to teachers in the same way.
The uncomfortable question is: what’s the difference?
Part of the answer is that portfolios aren’t measuring the same thing as a student essay. They’re measuring a career, not a moment. The evidence of teaching is in the classroom, not in the document. If the document is cleaner because of AI, but the teaching it describes was real, is the portfolio dishonest?
There’s no clean answer. Most teachers already know that — it’s the same murky territory they’ve been navigating on the student side for two years. The institutional conversation just needs to catch up. The ethical standard shouldn’t only flow downhill.
The Rec Letter Problem
University admissions offices have been consistent on one thing: the personal statement should be the student’s own work. The Common App says so directly. Most selective schools have added explicit language about AI use in application essays. The message is clear — the essay is supposed to be theirs.
Teachers, meanwhile, are writing hundreds of recommendation letters every fall. AI makes that faster. Drop in some notes about a student, get a polished draft in thirty seconds. It reads well. It hits the right notes about character and potential. And almost no one is asking whether that’s appropriate.
The asymmetry is worth sitting with. The student’s personal statement is supposed to represent their authentic thinking. So is the teacher’s letter — colleges use rec letters specifically because they want a human assessment from someone who actually knows the student. One document is held to an explicit standard about AI. The other is barely mentioned.
There’s a fair argument that it’s not equivalent — that a rec letter is a professional document, that using AI to help organize your thoughts isn’t fundamentally different from any other writing tool. That argument deserves a hearing. But so does the other side: if the letter was drafted by a language model from a few bullet points, how much of the assessment is actually yours? And is that what the college was asking for?
The larger point isn’t whether teachers should or shouldn’t use AI to help write rec letters. It’s that we’re enforcing a standard for students that we haven’t clearly articulated for ourselves. The same students whose college essays we’re reviewing are watching how we handle that.
A Harder Complication
The same week EdWeek ran that piece, Microsoft published its Q1 2026 Global AI Diffusion Report. The headline number: 17.8% of the world’s working-age population now uses generative AI, up from 16.3% six months ago. Progress.
The subtext is less encouraging. In the Global North, 27.5% of working-age people use AI. In the Global South, 15.4%. That gap widened in the past six months — it didn’t narrow. Language is one of the main barriers: AI tools work significantly better in English and a handful of dominant languages than in the hundreds of languages spoken across most of the world. Microsoft flagged this directly, calling the growing divide an urgent equity problem.
What that means for the integrity debate: the conversation we’re having — should teachers use AI for portfolios, should students use AI for essays, where exactly is the line — is a conversation happening in a narrow slice of the world. A slice where the tools are accessible, where the question is genuinely “should I?” rather than “can I?”
Most of the world’s teachers and students are not having this conversation. The debate about responsible AI use assumes access to AI in the first place. And while schools in Boston are mandating AI literacy as a graduation requirement — the first major US city to do so — the Microsoft data is a reminder that “AI literacy” means something very different depending on where you were born.
That’s not an argument to stop having the conversation here. It’s an argument for remembering what the conversation is actually about, and who it leaves out.
What This Means in a Classroom
The portfolio story matters for practical reasons, not just philosophical ones. If teachers are using AI to write their professional reflections — and students know it, or suspect it — the authority of the teacher’s position in that conversation shifts. You can’t tell a student that AI use is a crutch and then quietly use it to write the documents that certify your competence. Not coherently.
That’s not a gotcha. It’s a call for consistency.
The most defensible position — for teachers and students — is probably the same one: be clear about what AI did and what you did. Not as a confession, but as a practice. If you used AI to organize your thinking and then wrote the final document yourself, that’s one thing. If you prompted AI to write the reflection and edited lightly, that’s another. The difference isn’t binary, and neither is what we’re asking students to do.
What we’re really asking, in both cases, is: where did the thinking happen? And does the document tell the truth about that?
The standard applies on both sides of the desk.
Takeaway for Teachers
Before you finalize your next portfolio entry, rubric comment, or formal written self-assessment — anything you’d normally do yourself — ask the question you’d ask a student: what did AI do here, and what did I do? Not to police yourself, but to get clear on your own standard. Whatever answer you land on, let it inform what you actually say to students about their work. Consistency is the thing. Not perfection.
