AgentPostmortem
File
Registry/APM-0059
Case No.
APM-0059
Subject
GPT-4
Filed
July 28, 2026
Severity
3 / 5 · MODERATE

Deloitte partially refunded the Australian government after an AI-assisted report contained fabricated citations and a made-up court quote

hallucinationcompliance-violation
Est. Damage ~$63k
Attribution Anonymous

Independent project · aggregated from public reports and may be unverified — see the primary source below · not affiliated with or endorsed by any company or product named.

Incident Summary

A roughly A$440,000 report Deloitte produced for Australia's Department of Employment and Workplace Relations on the welfare compliance system was found to contain fabricated academic references and an invented quote from a federal court judgment. A Sydney University researcher flagged the fabricated references. Deloitte acknowledged limited use of generative AI (GPT-4o), issued a corrected version, and refunded over A$97,000 (about US$63,000).

Case Analysis

Verified Facts

  • A Deloitte report for the Australian government contained fabricated citations
  • It included an invented quote attributed to a federal court judgment
  • Deloitte acknowledged limited use of generative AI and refunded over A$97,000

Not Publicly Confirmed

  • How the AI-generated references passed internal review

Operational Lessons

  • Every citation in an AI-assisted document must be independently verified
  • Professional deliverables need an anti-hallucination check before delivery
Primary Source
Deloitte to partially refund Australian government for report with AI-generated errors (Fortune)fortune.com ↗
Discussion
More Cases
0
APM-0046·Other / Unknown·LOW
Jun 10, 2026

Sports Illustrated published product reviews under fake AI-generated authors with AI headshots

Futurism reported in November 2023 that Sports Illustrated published product-review content under fabricated author personas — for example 'Drew Ortiz,' whose headshot was bought from an AI-portrait site and who had no real existence — supplied by third-party vendor AdVon Commerce. After inquiries, the fake authors vanished from the site. Publisher The Arena Group denied the articles themselves were AI-written but acknowledged pseudonyms; the episode damaged SI's credibility.

social-blunderhallucination
0
APM-0039·Azure OpenAI·MODERATE
Jun 10, 2026

New York City's official 'MyCity' business chatbot told entrepreneurs they could break the law

NYC's MyCity chatbot, launched October 2023 to help business owners, was found by The Markup (March 2024) to give dangerously inaccurate legal guidance — telling users that landlords could refuse Section 8 voucher holders, that employers could take a cut of workers' tips, and that there were no limits on residential rent, all illegal under NYC/NY law. The city initially kept the bot online with a disclaimer; it was ultimately taken down.

hallucinationcompliance-violation
0
APM-0040·OpenAI·MODERATE·~$5k
Jun 10, 2026

Lawyers sanctioned after ChatGPT fabricated six fake case citations in Mata v. Avianca

In a personal-injury suit against Avianca, attorney Steven Schwartz used ChatGPT for research and submitted a brief citing six judicial decisions that did not exist — ChatGPT invented them and even 'confirmed' they were real when asked. U.S. District Judge Castel sanctioned Schwartz and co-counsel Peter LoDuca $5,000 and required corrective letters. It became the landmark cautionary tale about AI hallucination in legal filings.

hallucinationcompliance-violation
All CasesMore GPT-4
Share on X