Key Takeaways
- Farnsley Middle School in Kentucky distributed a student agenda riddled with AI‑generated gibberish—misspelled states, a bogus planet, and a corrupted periodic table.
- Local news outlet WDRB first reported the story, noting the school district offered no clear explanation beyond “communicating with parents about the issue.”
- Tests of leading chatbots (ChatGPT, Grok, Claude) produced accurate maps and tables; Google’s Gemini, however, created similarly nonsensical U.S. maps when prompted.
- The district’s website claims it sees generative AI as a tool to augment teaching, yet the error‑filled materials suggest AI was used as a substitute for human review.
- Additional AI‑generated images on the district site (extra‑fingered kids, floating tablets) hint at a broader pattern of unchecked AI output.
- The district has not responded to press inquiries, leaving open questions about which specific AI tool was used and why safeguards failed.
Overview of the Incident
Farnsley Middle School, part of Jefferson County Public Schools in Louisville, Kentucky, recently handed out a bound student agenda that contained a litany of obvious mistakes. As reported by WDRB, “the periodic table was nonsense. The map of the U.S. lists the state of Illinois as ‘Vitoiis,’ and Kentucky is listed as ‘Venecky.’ There’s an extra planet called Marc, and Saturn is listed as ‘Sature,’ among countless other gibberish errors.” The document was clearly not vetted before distribution, prompting disbelief among parents, educators, and online observers.
Details of the AI‑Generated Errors
Beyond the mangled map and periodic table, the agenda featured fabricated scientific facts that would be instantly recognizable as false to anyone with a basic education. The spurious planet “Marc” appeared alongside the real eight planets, while the altered periodic table misplaced elements and omitted groups entirely. Such errors are characteristic of low‑quality generative AI outputs when the model lacks sufficient grounding in factual data or when prompts are poorly constructed. The sheer volume of mistakes suggests the content was generated wholesale and then pasted into the agenda without any human curation.
Media Coverage and Initial Reaction
WDRB’s video segment highlighted the absurdity, showing side‑by‑side comparisons of the erroneous agenda pages with correct reference materials. The outlet noted that the school district “didn’t receive a proper explanation … except to say that ‘they are communicating with parents about the issue,’ nor did they get an answer about whether this was isolated to one school or the entire district.” The lack of transparency fueled speculation on social media, with many questioning how such blatant inaccuracies could reach students in the first place.
Testing Major AI Chatbots
To determine which AI system might have produced the flawed graphics, the authors experimented with the three largest publicly available chatbots: OpenAI’s ChatGPT, xAI’s Grok, and Anthropic’s Claude. Each generated a reasonably accurate map of the United States, correctly labeling states (Claude used state abbreviations rather than full names). Likewise, the periodic tables produced by these models were chemically sound, showing proper element symbols, atomic numbers, and groupings. This outcome indicated that the mainstream models, when prompted appropriately, are capable of delivering reliable educational content.
Suspicions Around Google Gemini
When the same experiment was repeated with Google’s Gemini—specifically its image‑generation component referred to as “Nano Banana”—the results diverged sharply. “A map of the U.S. labeled with gibberish nonsense, created using Google’s Gemini on Aug. 18, 2026,” the article shows, reproducing the kind of absurd state names seen in the school agenda (e.g., “Vitoiis” for Illinois, “Venecky” for Kentucky). Gemini’s tendency to hallucinate geographic labels under certain prompt conditions appears to match the observed errors, raising the possibility that the school used Gemini—or a similar image‑generation model—to create the agenda graphics.
Possibility of Legacy or Obscure Tools
The article also considers alternative explanations. The errors could stem from an older AI image generator that was prevalent a year or two earlier, when such hallucinations were more common. Alternatively, the district might have licensed a niche, proprietary AI tool marketed specifically for educational material production. The district’s website includes a section on AI that states, “We view GenAI not as a replacement for human intellect or the invaluable role of educators, but as a powerful tool to enhance learning, teaching, and operational efficiencies.” If such a tool were in use, the failure to apply human oversight would contradict the stated policy.
District’s Stated AI Policy vs. Practice
Jefferson County Public Schools’ public stance emphasizes AI as an augmentative resource, not a replacement for teacher judgment. Yet the agenda incident suggests the opposite: the treated AI output as final copy, bypassing any fact‑checking or editorial review. As the piece observes, “Even if AI spits out something this horrible, you can’t really blame the AI when something like this reaches students. It’s a bit like trying to blame Google if you found something on a sketchy website while doing a basic search and passing it directly to students without fact‑checking any of it.” The responsibility, therefore, lies with the adult who approved the material for distribution.
Broader AI‑Generated Content on District Website
Further evidence of unchecked AI use appears elsewhere on the district’s online presence. The article points out “AI slop featuring kids with too many fingers, phones with nonsensical designs, and tablets that hover in front of students.” One image shows a magically floating tablet on the Jefferson County Public Schools website, underscoring a pattern where AI‑generated visuals are published without correction. This systemic lapse indicates that the agenda error may not be an isolated mistake but a symptom of a wider workflow that trusts AI output implicitly.
Lack of Transparency and Ongoing Questions
As of the article’s publication, the school district had not replied to emailed inquiries sent on Tuesday afternoon. Gizmodo noted it would update the story if a response arrived and invited readers to share any knowledge about the specific AI chatbot responsible. The silence fuels speculation: is the district avoiding embarrassment, uncertain about the tool’s provenance, or concerned about liability? Until clarification is provided, educators, parents, and policymakers are left to grapple with the implications of deploying generative AI in classrooms without adequate safeguards.
Implications for AI Use in Education
The Farnsley Middle School episode serves as a cautionary tale for school systems embracing AI. While generative models can streamline content creation, they are not infallible; hallucinations, factual drift, and visual absurdities remain real risks. Effective integration demands robust human oversight—teachers or staff must verify AI‑generated material before it reaches learners. Moreover, districts should maintain transparency about which AI tools they employ, establish clear approval pipelines, and provide training on recognizing AI‑produced errors. Only then can the promise of AI as a “powerful tool to enhance learning” be realized without compromising educational integrity.
https://gizmodo.com/which-ai-chatbot-did-this-school-use-to-make-such-a-ridiculous-map-2000800045