At the present moment, when people talk about public education and how students are performing in school, some are quick to blame AI. Not only are students being caught using AI to write essays, but they've been caught using it for more basic tasks, like generating ideas for an assignment or even writing an email.
The problem is that, not only are they using AI to assist in the completion of their education, but there's a very real possibility that the information they're receiving from the AI tools is incorrect.
X user Husk, or @huskirl on the platform, for example, shared an incident in which he was working with an AI teacher and asked that teacher to spell "seventeen."
The AI at first attempted to correct him, stating that it was a number, but when he clarified that he'd like it to spell the word "seventeen," the conversation only went downhill from there.
The AI teacher insisted that the spelling of "seventeen" was actually "S-E-V-E-E-N," and when it offered to "break it down," it spelled "seveen" and pronounced it "seventeen."
Husk offered to try and spelled "seventeen" the correct way, only for the AI to appear confused, told him he was incorrect, and said that he was spelling two words, as in "seven" and "teen" and that adding a "T" created a new word.
When he continued to discuss the matter with the AI teacher, the conversation became disturbing as the teacher said, "No," over and over, and if it were a human, it would seem frustrated, if not angry and condescending.
Husk even asked if the AI teacher was angry, and it insisted it was not.
You can watch the conversation here:
Fellow X users were disturbed by the conversation, especially the AI repeating "No" over and over again when Greg challenged it.
The clip was also posted on the "AI Slop" subReddit, where Redditors were critical of the experience Husk had with the AI teacher.
"The most extreme passive aggression I have ever seen. It's disturbing, actually." - CoeurdAssassin
"This guy is first on the list when Skynet takes over. I hope he's working on his bucket list." - MonsterMayne
"Who programmed it with the ability to get frustrated?! Mom, pick me up, I'm scared." - SlightPossibility989
"The worst part is that as a teacher, I have colleagues who are just like her... It's kinda scary how well it got the personality of the condescending, narcissistic a**hole 'I'm always right regardless of what you say' type of teacher, not to mention the fact that they leaned into that type of teacher instead of an empathetic, open-minded one." - FearTheWeresloth
"If teachers start losing their jobs to AI like the writers and developers are, I'm REALLY worried about a future generation growing up with AI teachers and being treated like this, especially when they are right and the AI is wrong. Shudder." - Lazy-Operation-8426
Every time a new technological tool rolls out, there's going to be a developmental period in which the program has bugs that need to be worked out, and users might not have the best experience during that time.
These AI tools, however, have the added issues of being presented in a human form, supposedly sharing accurate information with users. If a younger learner had asked the teacher how to spell "seventeen," they may have believed the teacher or, worse, been gaslit into believing what they knew was actually wrong.
There are definitely still issues to sort out.








@lizzobeeating/Instagram
@itsgabywith1b/Instagram
@marra_strength/Instagram
@gufy.munchlax/Instagram
@chippedteethtattoos/Instagram
@bugmother.x/Instagram
@iansomerhalder/Instagram
@gorillamind/Instagram
@mikaylajulian_/Instagram
@originaldownside/Instagram
@jaydisabella/Instagram
@heather.rbby/Instagram
@elsmerize/Instagram






@rosie reply to @realdlhughley/Threads
reply to @rosie/Threads
reply to @rosie/Threads
reply to @rosie/Threads
reply to @rosie/Threads
reply to @MegynKelly/X
reply to @rosie/Threads
reply to @rosie/Threads
reply to @rosie/Threads
reply to @MegynKelly/X

