AI failures · Fabricated content

Ars Technica Retracts Article Containing AI-Fabricated Quotations Attributed to Scott Shambaugh

Verified Checked against its sources on October 20, 2026 4 sources quoted

Where it happened: United States
Organization: Ars Technica (Condé Nast)
Date: 2026-02-13 (published 2:40 p.m. EST, removed 4:22 p.m. EST the same day); editor's note retracting it 2026-02-15
Function: Journalism
Where it happened: Article writing; quotation sourcing
System: AI tools used while extracting quotations (reporter named an experimental Claude Code-based tool and ChatGPT)
Technology: Generative AI
Type of failure: Fabricated content
Caught by: The person affected, A journalist

Summary

On February 13, 2026, Ars Technica published an article about an AI agent that had written a hit piece on open-source maintainer Scott Shambaugh (see fail-045). The article attributed to Shambaugh quotations he had not said or written. One of its reporters said he had used AI tools while pulling material from Shambaugh's blog, ended up with a paraphrase presented as quotes, and asked Ars to pull the story once he realized the mistake. Shambaugh said publicly that the quotes were not his. Ars removed the article the same afternoon, and on February 15 published an editor's note retracting it and apologizing to him. TheWrap later reported that Ars dismissed the reporter.

Consequence

Article retracted and replaced by an editor's note apologizing to readers and to the person misquoted
Reporter dismissed by Ars Technica

Who caught it The person affected, A journalistFound outside the organization

Analysis

What could have been done

Checking AI output against its sources

Every factual element of AI output (citations, figures, quotations) is checked against its source before reliance.

Quotations that a reporter pulled from the source's own blog with AI tools were published as the source's words, although the source had not said or written them; the blog they were drawn from was available to check against.

A review that can actually catch the error

The reviewer has the time, information and expertise to catch the error, not only to sign off.

The fabricated quotations reached publication; the record shows no pre-publication check that caught them. The article was removed the same day, and the person misquoted said publicly the quotes were not his. The reporter said he had asked Ars to pull the story, and an editor's note retracted it two days later.

What would likely have prevented it

Treat any wording extracted or condensed by an AI tool as paraphrase until someone has matched it word for word against the original text or recording.
Require every direct quotation in an article to carry a pointer to its original location, so an editor can open the source and confirm the exact words before publication.

What would likely have caught it earlier

Have an editor check a sample of quotations against their linked sources before publication, starting with articles where AI tools were used to gather material.

Key dates

February 13, 2026Event
October 16, 2026Checked against sources
October 18, 2026Checked against sources
October 20, 2026Checked against sources
October 21, 2026Checked against sources

Developments are added when the record is reviewed. Next review: October 20, 2027.

Evidence

Each passage is quoted from its source. The quoted passages are re-checked against the live pages every month.

On Friday afternoon, Ars Technica published an article containing fabricated quotations generated by an AI tool and attributed to a source who did not say them.

Supports: what happened; publisher's own retraction

Editor’s Note: Retraction of article containing fabricated quotations Ars Technica (Ken Fisher, Editor in Chief) Primary source
Shambaugh updated his blog to say he did not talk to Ars Technica and did not say or write the quotes in the articles.

Supports: detection: the person misquoted said publicly he never said the quotes

The Ars Technica article, which had two bylines, was pulled entirely later that Friday.

Supports: removal on the day of publication (corroborates source 4)

Ars Technica pulls article with AI-fabricated quotes about AI-generated article 404 Media (February 15, 2026) Reporting or analysis
Ars Technica, the Condé Nast-owned technology outlet, fired senior AI reporter Benj Edwards after the publication retracted one of his stories over the use of fabricated quotes.

Supports: consequence: reporter dismissed

Edwards’ account of the Shambaugh incident was published on Feb. 13; it was taken down two days later and replaced with an editor’s note by editor-in-chief Ken Fisher that apologized for the story.

Supports: editor's note by Ken Fisher apologizing for the story; TheWrap's 'taken down two days later' is contradicted by Ars's own notice (source 4), which gives removal on February 13

After realizing the mistake, Edwards asked the outlet to pull the report.

Supports: the reporter's account that he asked Ars to pull the story

Ars Technica Fires Reporter Over AI-Generated Quotes TheWrap Reporting or analysis
Ars Technica has retracted this article. Originally published on Feb 13, 2026 at 2:40PM EST and removed on Feb 13, 2026 at 4:22PM EST.

Supports: dates: published and removed on February 13, 2026 (the publisher's own notice)

Retraction: After a routine code rejection, an AI agent published a hit piece on someone by name Ars Technica (retraction notice at the original article URL) Primary source

Suggested citation. Eminence Scan, “Ars Technica Retracts Article Containing AI-Fabricated Quotations Attributed to Scott Shambaugh”, Eminence Growth Solutions, last verified October 20, 2026, https://eminencegrowthsolutions.com/tools/eminence-scan/record/fail-037/.

Using our research: you may read and quote Eminence Scan with credit to “Eminence Scan, Eminence Growth Solutions” and a link; all other rights reserved. Terms: Using our research.

Spotted an error, or a development we have missed? Report a correction. How we check records: research methodology.