Something is happening, and I cannot stop thinking about it.
Over the past several months, artificial intelligence has moved from a specialized topic discussed by researchers and Silicon Valley insiders to front-page news that my students' parents are asking me about at conferences. DALL-E 2 can generate photorealistic images from text descriptions. GPT-3 can write essays that are, to an untrained eye, indistinguishable from student work. Stable Diffusion has made AI image generation available to anyone with a computer. GitHub Copilot is writing code alongside professional developers.
I am not an AI researcher. I am a K-12 educator with 28 years of classroom experience, a Google Certified Educator who earned both Level 1 and Level 2 credentials this past summer, and an Apple Teacher who has spent her career integrating technology into instruction. I do not have technical expertise in machine learning or neural networks. But I have something that I think is equally valuable right now: a practitioner's instinct for when a technology is about to fundamentally change the terms of my profession.
That instinct is sounding alarms.
What I Am Observing
Let me describe what I am seeing at ground level, in a real school, with real students.
Two weeks ago, a student showed me an image he had generated using a free AI art tool. It was a photorealistic portrait of a person who does not exist, created in seconds from a text prompt. He was delighted. I was unsettled - not by the image itself, but by the casual ease with which it was produced and the complete absence of any framework for thinking about what it meant.
Last month, a colleague in the English department told me that a student submitted an essay that was "suspiciously good" - syntactically polished, structurally sound, but somehow lacking the voice and rough edges that characterize genuine student writing. She could not prove anything. She could not even articulate precisely what felt wrong. She just knew.
These are early signals, and I am paying attention.
The Capabilities Are Real
I have spent the last few weeks deliberately exploring the AI tools that are making headlines, and I want to be honest about what I found: the capabilities are not hype. They are real, they are accessible, and they are improving at a rate that makes linear planning obsolete.
GPT-3, developed by OpenAI, is a language model that can generate coherent, contextually appropriate text in response to prompts. I tested it myself, giving it prompts similar to writing assignments I give my students. The output was not perfect - it was occasionally generic, sometimes factually uncertain, and lacking the specific personal experience that distinguishes strong student writing from competent but empty prose. But it was good enough to pass. That is the threshold that matters.
DALL-E 2 generates images from text descriptions with a level of quality that, even six months ago, would have seemed like science fiction. The implications for visual arts education, media literacy, and digital citizenship are profound and largely unexamined.
And these are the tools available today. The trajectory suggests that what seems impressive now will seem primitive within a year or two.
Why Education Should Be Paying Attention Now
The education sector has a well-documented pattern of responding to technological disruption reactively rather than proactively. We waited until students were already plagiarizing from the internet before developing digital literacy curricula. We waited until cyberbullying was a crisis before integrating digital citizenship into school culture. We waited until the pandemic forced our hand before meaningfully investing in remote learning infrastructure.
We cannot afford to wait on AI.
The reason is speed. Previous technological disruptions gave education years - sometimes a decade - to observe, discuss, pilot, and adapt. The internet was commercially available in the mid-1990s; schools did not seriously grapple with its implications for instruction until the mid-2000s. Social media emerged in the mid-2000s; school policies began catching up around 2010-2015.
AI is moving faster. The gap between "interesting research demo" and "consumer product that any student can access for free" has compressed from years to months. By the time a committee convenes to discuss implications, the technology will have already reshaped the landscape the committee is examining.
The Questions We Need to Ask
I do not pretend to have answers. But I have been developing a set of questions that I believe every educator needs to start wrestling with, even if the wrestling is uncomfortable.
What does "original student work" mean when AI can generate text, images, and code? Our entire assessment infrastructure is built on the assumption that submitted work reflects individual student capability. That assumption is becoming unstable. This is not simply a plagiarism problem - it is an epistemological crisis about what we are actually measuring when we grade student work.
What skills become more valuable, not less, in an AI-augmented world? If AI can generate a competent five-paragraph essay, then the ability to write a competent five-paragraph essay may not be the differentiating skill we have treated it as. Critical thinking, ethical reasoning, creative problem-framing, emotional intelligence, and the ability to evaluate and improve AI-generated output - these may be the capabilities that matter most. Are we teaching them?
How do we maintain equity when AI tools have varying costs and access barriers? The digital divide that characterized the pandemic - students with reliable devices and internet versus those without - will be replicated and amplified if AI tools become essential to academic success while remaining unequally distributed.
What are the ethical boundaries, and who draws them? Should students be allowed to use AI writing tools as part of their process? Under what conditions? With what disclosure requirements? These are not rhetorical questions. They require actual policy, and that policy should be developed by educators, not imposed by technology companies.
What I Am Doing About It
At the individual level, my response has three components.
First, I am educating myself. I have added AI literacy to my ongoing professional development alongside my Google certification work. I am reading broadly - not just the breathless technology coverage, but the thoughtful criticism, the ethical analyses, and the educational research that is just beginning to emerge. I want to understand what these tools actually do, not just what they appear to do.
Second, I am talking to my students. Not formal lessons yet - I do not know enough to teach this with the rigor it deserves - but open conversations about what they are seeing, what they are using, and what they think about it. Their perspectives are essential data that I need before forming my own positions.
Third, I am talking to my colleagues. In faculty meetings, in hallway conversations, in our department Slack channel, I am raising the topic and gauging where people are. The range is enormous - from teachers who have not heard of GPT-3 to teachers who are already using AI tools to generate lesson plan outlines. We need to get on the same page before we can move forward coherently.
The Feeling in My Gut
I want to end with something less analytical and more honest. I have been teaching for nearly three decades. I have lived through the introduction of the internet into schools, the 1:1 laptop initiative, the interactive whiteboard era, the LMS revolution, the app explosion, and the pandemic's forced digitization. Each of those shifts changed some aspect of my practice.
What I am sensing with AI feels different - not in degree but in kind. The previous technologies changed how we deliver instruction. AI has the potential to change what instruction means. That is a fundamentally different category of disruption, and it requires a fundamentally different level of preparation.
I do not know what education looks like in five years. But I am increasingly certain that the decisions educators make in the next 12 to 18 months - about how we understand AI, how we teach students to use it ethically, and how we redesign assessment for a world where machines can produce human-quality output - will shape the profession for a generation.
The time to start is now. Not when the tools are more mature. Not when the research is conclusive. Not when someone hands us a curriculum guide. Now, while we still have the opportunity to be proactive rather than reactive for once in education's history.
I will be writing more about this in the months ahead. Something tells me there will be no shortage of material.
Dr. Janette Camacho is a veteran K-12 educator with 28+ years of classroom experience, a Google Certified Educator Level 1 and Level 2, and an Apple Teacher. She writes about the intersection of pedagogy and technology at iTeachAI.