The Messenger signs a deal to “eliminate bias” and flag “clickbait” with AI
What counts as clickbait? Merriam-Webster’s definition is “something (such as a headline) designed to make readers want to click on a hyperlink especially when the link leads to content of dubious value or interest.”
Keep that in mind as you browse some of the top headlines on The Messenger’s homepage this morning:
- “Ed Sheeran Says Snoop Dogg Got Him So High He ‘Couldn’t See'”
- “Late-Night Hosts Roast Kevin McCarthy: ‘You Are History'”
- “Trump Drops Off Forbes List of Richest Americans”
- “Rep. Chip Roy on Gaetz RINO Comments: ‘You Can Kiss My A–.'”
That line-up is just one reason why people were just a little confused to see The Messenger — which launched earlier this year — announce they’d partnered with AI company Seekr to flag “clickbait” and “eliminate bias” in their reporting.
The press release claims “the partnership makes The Messenger among the first media publishing companies to leverage AI to ensure editorial content consistently aligns with journalism standards.”
The Messenger says it is partnering with AI company Seekr because humans are bringing too much personal bias into journalism pic.twitter.com/vY2KW9ZGvr
— Max Tani (@maxwelltani) October 4, 2023
85% of The Messenger’s content output now is just rewriting news already produced by other outlets, usually without attribution. So now they’ll just use *Artificial Intelligence* to replace humans to scavenge, appropriate, regurgitate with less “personal bias”
— jim haigh (@jmhaigh) October 4, 2023
For now, The Messenger reporters will not be required to use the tool. Those who opt-in will “receive a score reflecting their political lean and reliability that will be visible to readers,” Adweek reported.
We’ve seen a lot of money thrown at projects that promise to “solve bias” once and for all. Seekr seems even less familiar with journalism than some of these other efforts. One giveaway? It calls headlines “titles” throughout its site. Its “clickbait” score, for example, is based on whether “the title appeal to curiosity or emotion instead of describing the story.”
“We believe that in the hands of publishers, our technology will be an extraordinarily effective quality control tool,” Rob Clark, president and chief technology officer at Seekr, says in the release. “The technology is accurate, reliable, and extremely fast, and it can help ensure that published content consistently adheres to basic standards.”
After taking a look at which articles have earned “very high reliability” scores from Seekr, we only have more questions.
I just went to Seekr’s website where they rate an OANN article as “very high reliability” and “centrist,” and call the Daily Mail “left center.” The AP, meanwhile, is “low” reliability. Worse than useless, actively misleading. https://t.co/gUHiy6xylD pic.twitter.com/PRsDzt0cAZ
— Christopher Ingraham (@_cingraham) October 4, 2023
For giggles, I checked how this AI company @Seekr_io currently rates the “reliability” of stories on Kevin McCarthy’s ouster as speaker.
VERY HIGH: Breitbart, Daily Mail, NY Post, OANN, ZeroHedge, WashTimes
VERY LOW: Bloomberg, NYT, BBC, Guardian, ABC, NBC, Time, Politico
https://t.co/3kY1CJnDCX pic.twitter.com/QDu3ewts73
— Joshua Benton (@jbenton) October 4, 2023
Meanwhile, some “VERY LOW” reliability stories: pic.twitter.com/WEzvC1IVpq
— Joshua Benton (@jbenton) October 4, 2023
Yes, I think any podcast company that partners with Seekr for brand safety should really really re-consider. The Dan Bongino Show has a high civility score. Sure. https://t.co/LU2pOb7zwU
— Amrita Khalid (@askhalid) October 4, 2023
Honestly the list on the left is mostly a who’s who of misinfo spreading sites (plus most of the rest are extremely partisan sites). If this is the system The Messenger thinks will deliver journalistic rigour and impartiality, I have a whole portfolio of bridges to sell them. https://t.co/6sXtCdpHZ8
— Jasper Jackson (@JaspJackson) October 4, 2023