Students forgetting to delete AI chat prompts from school essays
Students are turning in school papers that still contain conversational filler and wrong topics generated by AI tools because they fail to review their work before submission. This accidental oversight leaves obvious chat remnants like introductory responses in the final text, immediately exposing them to academic penalties.
What the numbers show
Counted from the threads quoted below. How we count
- It goes quiet, then comes back — 3 separate clusters since October 2025.
What people tried
Every workaround mentioned in the threads below. We haven’t tested any of them — and nobody here is claiming they worked.
Found another way around this?
Share your steps and trade-offs so fellow engineers don't have to reinvent the wheel. No account required.
- 1Making students write all of their papers by hand using pencil and paper in classDid this help you?
- 2Placing hidden text prompts in assignments to catch students who copy and paste without readingDid this help you?
- 3Pulling students aside to ask why their papers contain identical weird or unrelated contentDid this help you?
- 4Checking for uncharacteristic grammar, diction, and dead giveaways like pasted AI error messagesDid this help you?
- 5Encouraging students to start a petition or protestDid this help you?
- 6Going directly to administration to complainDid this help you?
- 7Asking a tenured colleague to approach administrationDid this help you?
- 8Pushing for rule changes to ban artificial intelligence in future contestsDid this help you?
- 9Asking students to explain their writing in their own wordsDid this help you?
- 10Using alternative checkers like PangramDid this help you?
- 11Using document version history tools like DraftbackDid this help you?
- 12returning to paper and pencil assignments in classDid this help you?
- 13doing everything in class and on paperDid this help you?
- 14eliminating technology useDid this help you?
- 15Using lockdown browsers and proctoring services for assessmentsDid this help you?
- 16Checking Google Docs keystrokes and edit history using a scrubberDid this help you?
- 17Pulling suspected students into breakout rooms to ask them about their workDid this help you?
- 18Introducing fake chemical elements or concepts to trap AI chatbotsDid this help you?
- 19Moving to purely presentation-style assessmentsDid this help you?
- 20Giving zeroes directly and contacting parentsDid this help you?
In their words
Unedited, grouped by where they were said, most upvoted first within each place, each linked to the thread it came from.
“One of the seniors turned in an English paper about reducing homeliness when the paper prompt was about reducing homelessness.”source ↗
“What baffles me about this is that these kids aren't even reading what AI has put out?”source ↗
“I tried using AI as a teacher to write a sample text once. And only once because even though I reread it carefully and made edits, it still had a major error the students caught.”source ↗
“They aren't using it for “help” - they don’t care what it says, just that they’re able to outsource thought in exchange for free time”source ↗
“I’m fuming that this happened.”source ↗
“I know that many others put their skills and efforts into hand drawing the design, and I’m mortified that a homeroom teacher submitted it and that admin chose the winner.”source ↗
“I’m thinking of putting together a petition to overrule this, but I don’t know if I’m overreacting.”source ↗
“First line “Sure! Here’s a narrative about xyz and 1200 words long!” When I showed him his error, I watched the blood drain from his face”source ↗
“I have personal bandwidth to fight this; I’m just not sure if I’ll be overstepping”source ↗
“I’m an overworked teacher, and I haven’t seen even one useful AND ethical thing AI can do for me.”source ↗
“When it comes to humanities and arts being generated by AI, I take it personally.”source ↗
“My admin *loves* AI! She brags all the time about how she uploads everything to AI and has it make her PowerPoints for staff development.”source ↗
“At a state university, I've colleagues who are now "customizing" or "updating" their curriculum and materials with LLM's. And then, the students are using AI to fullfill the course requirements.”source ↗
“So, essentially it is just AI talking to AI, with a person being an inefficient error prone buffer in-between.”source ↗
“My biggest battle involves students using AI for everything.”source ↗
“I struggle with not being able to 100% prove it.”source ↗
“How do I explain to my students that I would rather read their sucky, basic work than anything AI produces and work with them from that starting point.”source ↗
“Basically I want a way to tell them when I get AI generated work (you know you didn’t write this, I know you didn’t, and AI knows) without dragging parents into it every time.”source ↗
“I want to be approachable, realistic, and I really want to not spend time arguing with students/parents.”source ↗
“I deal with it daily myself.”source ↗
“I am somewhat in the same situation.”source ↗
Where this came up
People with this problem also raised
- 3School administration replacing student art with AI posters
- 4My students think all art and special effects are AI
- 44How to deal with all the low effort AI content online
- 25Why does fixing AI mistakes take longer than doing it myself?
- 4What tools detect cut-and-paste and LLM cheating on assignments?
- 6Why am I getting falsely accused of using AI to write?