The name “Stanford” has popped up in a slew of news reports in this country in recent days, as Education Minister Erica Stanford wriggles her way – or tries to – through various controversies.
She and the Principals’ Federation president are firing insults at each other after she delivered a speech at a recent federation conference.
She is embroiled in controversy over Official Information Act issues (here and here).
The Waitangi Tribunal (yawn – when will it ever give a government a clean bill of health?) has found Treaty of Waitangi breaches in the NCEA overhaul on Stanford’s watch.
But no, she has not used AI to replace a Latino student with a Black student in a promotional photo.
That bit of mischief was done at Stanford University, in the USA.
The headline above, from the Washington Times, draws attention to the use – or misuse – to which artificial intelligence can be put.
The report beneath the headline says:
Stanford University acknowledged this week that it used artificial intelligence to alter a promotional photo, swapping out a Latino student for an AI-generated Black woman and slimming down two of his classmates — the latest episode to raise questions about how universities market diversity, and one that lands amid mounting federal scrutiny of Stanford’s own record on race.
The doctored image was first flagged by the Stanford Review, an independent student publication, which published side-by-side comparisons showing Billy Ramirez and two classmates in a 2024 photo. When the image resurfaced on banners around campus this month, Mr. Ramirez had vanished, replaced by a woman of a different race and gender, while the two students beside him appeared visibly thinner than in the original shot.
Mr. Ramirez told the student paper the experience left him feeling silenced and erased from a representation that was supposed to include him.
Image credit: The Stanford Review
A Stanford spokeswoman has confirmed the alteration violated university policy, saying the school is working on improving staff training.
Stanford’s AI guidelines explicitly bar the creation of “synthetic media depicting real individuals” without consent and prohibit AI-generated content that “falsely depicts Stanford people” without disclosure.
The article notes this isn’t the first time a university has been caught digitally engineering the appearance of diversity.
In 2000, the University of Wisconsin at Madison admitted it had inserted a Black student into an all-white crowd photo for an admissions booklet cover, prompting officials to recall roughly half of the 110,000 copies they had printed — a reprinting effort contemporary news accounts put at about $64,000.
The byline on the report we have referenced reads:
By The Washington Times AI News Desk – Tuesday, September 22, 2026.
Here at PoO we don’t have an AI News Desk.
How many of New Zealand’s mainstream media (we wonder) have such a desk.
The end of the article is instructive, too. It reads:
This article was constructed with the assistance of artificial intelligence and published by a member of The Washington Times’ AI News Desk team. The contents of this report are based solely on The Washington Times’ original reporting, wire services, and/or other sources cited within the report. For more information, please read our AI policy or contact Steve Fink, Director of Artificial Intelligence, at sfink@washingtontimes.com
The Washington Times AI Ethics Newsroom Committee can be reached at aispotlight@washingtontimes.com.
Transparency.
It’s to be admired – and encouraged.
Besides musing on the merits of similar acknowledgements in our mainstream media, New Zealanders should take this stuff on board and be pressing for curbs on politicians and their party publicists as they gear up for the general election campaign.
AI-generated images have the potential to influence voter behaviour and there is a strong case for requiring political parties to acknowledge the use of AI in their advertising.
The Electoral Act regulates “election advertisements” broadly, enforcing spending caps and requiring “promoter statements” to identify who paid for an ad.
But New Zealand law features no general prohibition against publishing misleading or false election advertisements, which means the creation of deepfakes or AI-generated “smear” content falls into a regulatory gray area.
You can be sure party strategists will be working out how to make the most of those legal shortcomings.
Bob Edlin is a veteran journalist and editor for the Point of Order blog HERE.
But no, she has not used AI to replace a Latino student with a Black student in a promotional photo.
That bit of mischief was done at Stanford University, in the USA.
The headline above, from the Washington Times, draws attention to the use – or misuse – to which artificial intelligence can be put.
The report beneath the headline says:
Stanford University acknowledged this week that it used artificial intelligence to alter a promotional photo, swapping out a Latino student for an AI-generated Black woman and slimming down two of his classmates — the latest episode to raise questions about how universities market diversity, and one that lands amid mounting federal scrutiny of Stanford’s own record on race.
The doctored image was first flagged by the Stanford Review, an independent student publication, which published side-by-side comparisons showing Billy Ramirez and two classmates in a 2024 photo. When the image resurfaced on banners around campus this month, Mr. Ramirez had vanished, replaced by a woman of a different race and gender, while the two students beside him appeared visibly thinner than in the original shot.
Mr. Ramirez told the student paper the experience left him feeling silenced and erased from a representation that was supposed to include him.
Image credit: The Stanford Review
A Stanford spokeswoman has confirmed the alteration violated university policy, saying the school is working on improving staff training.
Stanford’s AI guidelines explicitly bar the creation of “synthetic media depicting real individuals” without consent and prohibit AI-generated content that “falsely depicts Stanford people” without disclosure.
The article notes this isn’t the first time a university has been caught digitally engineering the appearance of diversity.
In 2000, the University of Wisconsin at Madison admitted it had inserted a Black student into an all-white crowd photo for an admissions booklet cover, prompting officials to recall roughly half of the 110,000 copies they had printed — a reprinting effort contemporary news accounts put at about $64,000.
The byline on the report we have referenced reads:
By The Washington Times AI News Desk – Tuesday, September 22, 2026.
Here at PoO we don’t have an AI News Desk.
How many of New Zealand’s mainstream media (we wonder) have such a desk.
The end of the article is instructive, too. It reads:
This article was constructed with the assistance of artificial intelligence and published by a member of The Washington Times’ AI News Desk team. The contents of this report are based solely on The Washington Times’ original reporting, wire services, and/or other sources cited within the report. For more information, please read our AI policy or contact Steve Fink, Director of Artificial Intelligence, at sfink@washingtontimes.com
The Washington Times AI Ethics Newsroom Committee can be reached at aispotlight@washingtontimes.com.
Transparency.
It’s to be admired – and encouraged.
Besides musing on the merits of similar acknowledgements in our mainstream media, New Zealanders should take this stuff on board and be pressing for curbs on politicians and their party publicists as they gear up for the general election campaign.
AI-generated images have the potential to influence voter behaviour and there is a strong case for requiring political parties to acknowledge the use of AI in their advertising.
The Electoral Act regulates “election advertisements” broadly, enforcing spending caps and requiring “promoter statements” to identify who paid for an ad.
But New Zealand law features no general prohibition against publishing misleading or false election advertisements, which means the creation of deepfakes or AI-generated “smear” content falls into a regulatory gray area.
You can be sure party strategists will be working out how to make the most of those legal shortcomings.
Bob Edlin is a veteran journalist and editor for the Point of Order blog HERE.
No comments:
Post a Comment
Thank you for joining the discussion. Breaking Views welcomes respectful contributions that enrich the debate. Please ensure your comments are not defamatory, derogatory or disruptive. We appreciate your cooperation.