Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

AI news ·

Ethical AI push undermined as Newfoundland and Labrador education report cites at least 15 fake sources

N.L.'s 10-year education plan cites at least 15 fake sources, some likely AI-generated, eroding trust. Schools should mandate AI disclosure, verified DOIs, and citation audits.

3 min read
Education
Share

Fake citations found in Newfoundland and Labrador's 10-year education plan: what educators should do next

CBC News reports that Newfoundland and Labrador's 418-page education reform plan, "A Vision for the Future: Transforming and Modernizing Education," contains at least 15 fabricated citations. The document took 18 months to produce and was released August 28 by co-chairs Anne Burke and Karen Goodnough, alongside Education Minister Bernard Davis.

The problem is more than sloppy footnotes. Academics suspect some sources were generated by an AI language model, which clashes with the report's own recommendation to teach ethical AI, data privacy, and responsible technology use in schools.

What happened

One citation refers to a 2008 National Film Board movie titled "Schoolyard Games" that does not exist, according to a board spokesperson. Even more concerning, the exact citation reportedly appears in a University of Victoria style guide that uses fictional examples to teach formatting-clearly labeled as made up.

Aaron Tucker, an assistant professor at Memorial University who researches AI history in Canada, said he could not find multiple sources listed in the report across library systems, databases, and Google. "Fabricating sources is a telltale sign of artificial intelligence," he told CBC.

Josh Lepawsky, former president of the Memorial University Faculty Association, put it bluntly: errors are one thing; made-up citations demolish trust. Co-chair Karen Goodnough told CBC the team is investigating and checking references, and the Department of Education said the online report will be updated to fix errors.

Why this matters for schools and universities

Large language models produce text that looks correct before it is correct. That includes citations: they often fabricate, pick the wrong sources, or mischaracterize real ones.

When a public policy document carries fake references, credibility breaks. That damage spreads-to classrooms, to faculty, and to public confidence in your institution.

What you can do now

  • Set a clear AI-use disclosure rule: staff and students must indicate if AI tools contributed to research, drafting, or citations.
  • Require verifiable references: include DOI or stable URLs, publisher, year, and page numbers for quotes.
  • Adopt a reference verification checklist for all high-stakes documents (policy, accreditation, board reports).
  • Make source files mandatory: authors submit PDFs or screenshots for non-open sources and paywalled material.
  • Assign a "citations lead" on major projects to spot-check a random 20-30% of references before approval.
  • Involve librarians: add a formal citation audit pass for institutional documents.
  • Ban AI-generated citations: AI may help format, but all sources must be independently located and verified by a human.
  • Track quotes to pages: no quotation without a page number or timecode for audiovisual content.
  • Train staff on AI limits: plausible ≠ accurate; show examples of fabricated citations and how to validate them.

Suggested policy language you can copy

  • "Use of generative AI must be disclosed. AI tools may not invent or select sources. All citations must be verified by the author."
  • "Every citation must include a DOI or stable link when available. Quotes require page numbers or timecodes."
  • "Major reports undergo a citations audit by a designated reviewer or librarian before publication."
  • "Submitting unverifiable or fabricated sources is academic misconduct and grounds for correction or withdrawal."

A note on the AI problem itself

AI systems generate text by predicting what looks likely, not by checking truth. That is why they produce confident, formatted, and wrong references.

Even models that can search the web may still mix up authors, dates, or titles. Treat AI as a drafting assistant, not a source of truth.

Resources

Bottom line: credibility is a system. Build in AI disclosure, reference verification, and librarian review. Publish nothing high-stakes without a citations audit.

Share