Google AI Falsely Links Auckland Technician to Criminal Convictions
Generative AI creates damaging misinformation, threatening the professional reputation of an innocent healthcare worker.
1False Criminal Allegations
Tracey Melsom-Page, a respected anaesthetic technician with 39 years of experience, was falsely identified by Google’s AI as having drink-driving convictions. The AI linked her to a 2023 news story about a different practitioner, despite her name never appearing in the original report. This error has caused significant personal and professional distress.
2Systemic Algorithmic Failure
Google’s AI Overview initially doubled down on the false claims, asserting that it had cross-referenced official tribunal databases to verify the information. Only after repeated questioning did the system acknowledge a logical error. The technology incorrectly synthesized web content, creating a harmful narrative that did not exist in the source material.
3Professional Reputation Risks
Melsom-Page fears the AI-generated misinformation could permanently damage her career and standing among colleagues and future employers. Having spent nearly four decades building a reputation in public hospitals, she now faces the prospect of this false association appearing in search results for years, potentially impacting her future job prospects and credibility.
4Difficulty in Seeking Redress
The victim and her husband spent over three hours attempting to navigate Google’s online forms to remove the false content, but they encountered broken links and dead ends. Despite the severity of the harm, the couple found the process of contacting the technology giant directly to be an exhausting and largely unsuccessful endeavor.
5Broader Pattern of Errors
This incident follows a similar case involving New Zealand rich-lister Wayne Wright jnr, who was incorrectly linked to a child abuse case by Elon Musk’s AI chatbot, Grok. These instances highlight a growing trend where generative AI systems autonomously create and disseminate defamatory information, leaving innocent individuals struggling to clear their names.
6Expert Warnings on Safety
Sean Lyons, chief online safety officer at Netsafe, warned that AI systems are often deployed without sufficient safeguards to prevent such harm. He noted that AI can establish seemingly logical but entirely false connections between people and events, creating a significant challenge for individuals who find themselves targeted by automated misinformation campaigns.
7Legal Avenues for Protection
New Zealand’s Harmful Digital Communications Act provides potential legal avenues for those suffering emotional distress from digital content, regardless of whether it was generated by humans or AI. The District Court has previously ordered search engines to remove links, offering a pathway for victims to seek redress when technology companies fail to act.
8Corporate Accountability Gap
Google stated it is investigating the case and will take action to fix the issue, noting that most AI overviews are accurate. However, the company did not respond to requests regarding potential compensation or a formal apology. This highlights the ongoing tension between rapid AI deployment and the lack of accountability mechanisms.
Original article: Newstalk ZB — Google AI Falsely Links Auckland Technician to Criminal Convictions