- Last Thursday, The Irish Times removed an opinion article which called fake tan "problematic."
- It was 80% generated by GPT-4, and the hoaxer used an AI profile picture, per The Guardian.
- The editor has apologized and said it highlighted the challenges of generative AI for newspapers.
A hoax newspaper article which was generated with AI by a troll using a DALL-E profile picture has prompted an apology from the editor, The Guardian reported.
The Irish Times published the piece, titled "Irish women's obsession with fake tan is problematic," on Thursday. It became the second-most popular article on the site and sparked debates on radio and social media in the country.
"To me, fake tan represents more than just an innocuous cosmetic choice," the article said. "It raises questions of cultural appropriation and fetishisation of the high melanin content found in more pigmented people."
But it turned out that the contributor, Adriana Acosta-Cortez, wasn't actually an Ecuadorian health worker who'd moved to Dublin as stated. In fact, she never existed. The Irish Times removed the piece within 24 hours and began a review.
In a tweet, the hoaxer described the opinion piece as "divisive tripe" and the episode as "genuinely sad."
—Adriana Acosta-Cortez (@ecuadorian_adri) May 12, 2023
The person behind the account told The Guardian they're an Irish college student who identifies as non-binary.
About 80% of the article was generated with GPT-4, they told the newspaper. And the profile picture was created with DALL-E to show a stereotypically "woke" journalist with the prompt "female, overweight, blue hair, business casual clothing, smug expression."
"Some people have called me an alt-right troll but I don't think that I am," they told The Guardian. "I think that identity politics is an extremely unhelpful lens through which to interpret the world."
Since admitting to the hoax, the account retweeted a post from the far-left Revolutionary Housing League, speaking about the divide between homeless people and vacant homes.
On Sunday, The Irish Times editor, Ruadhán Mac Cormaic, apologized in a statement. "We had fallen victim to a deliberate and coordinated deception," he said.
Alongside "gaps" in its editorial procedures, "it has also underlined one of the challenges raised by generative AI for news organisations," he added. "We, like others, will learn and adapt."