Sports Illustrated Published Articles by Fake, AI-Generated Writers – Slashdot
Become a fan of Slashdot on Facebook
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
All so spontaneous and personal.
But we’re all craving for this kind of ‘information’.
From TFA
one Ortiz article, for instance, warns that volleyball “can be a little tricky to get into, especially without an actual ball to practice with.”
one Ortiz article, for instance, warns that volleyball “can be a little tricky to get into, especially without an actual ball to practice with.”
What was that you were saying about the articles being “informative, well-written, and enjoyable to read”?
I object to the use of AI “writers” on several grounds.
1) Using manufactured personas is dishonest. I don’t care about a plucky reporter’s origin story. But if I’m hearing a story by nobody, I’d like to know.
2) Contra today’s earlier article, AI is dis-employing writers.
3) It suggests that the owners of SI don’t really respect their readers, writers, or sports themselves, and provides a good example of Capital enshittifying everything it touches to get a little richer so it can buy more things to make shittier.
The article was cobbled together according to a set of rules, but conveys no human intent, idea, or POV (often useful in sportswriting. It is the prose-equivalent of wind-chimes- any melody or meaning is the result of efforts by the listener.
1) Using manufactured personas is dishonest.
1) Using manufactured personas is dishonest.
Sports writing has always been formulaic, with templates and story generators. This is just another step.
But if I’m hearing a story by nobody, I’d like to know.
But if I’m hearing a story by nobody, I’d like to know.
“Like to know” isn’t “right to know”. If you don’t like Sports Illustrated, don’t subscribe.
AI is dis-employing writers.
AI is dis-employing writers.
That is a GOOD THING. Automating work is the foundation of our prosperity. It is what separates the developing world from SH countries. People always object while it’s happening, but later universally agree it was a good thing. Nobody wants to abolish ATMs, self-dialing phones, and flush toilets despite all the un
“Like to know” isn’t “right to know”. If you don’t like Sports Illustrated, don’t subscribe.
“Like to know” isn’t “right to know”. If you don’t like Sports Illustrated, don’t subscribe.
The problem is GP could not know, and as you suggest this knowledge would have changed the decision to purchase. Which is what makes it a fraud. The amount is too inconsequential to sue, of course.
developing world from SH countries.
developing world from SH countries.
South Heurope?
AI-generated content has won awards from judges who didn’t know it was AI-generated, so the objective evidence is that it can be just as creative as humans, despite your “feelings”.
AI-generated content has won awards from judges who didn’t know it was AI-generated, so the objective evidence is that it can be just as creative as humans, despite your “feelings”.
There is also a story that random garbage that was not supposed to be there also won Arts Awards. The whole judgement of creative pieces relies on the common understanding of the creation process and the skills needed to achieve the result. If you just judge by the result, an inkjet printed photograp
Before jumping to conclusions about the dangers of AI writers, I’d first like to better understand the actual reason that the writers in this case felt the need to use pseudo names.
It’s going to be quite a pathetic argument if the reason buried in “author privacy” stems from the current cancel culture that can voraciously attack and destroy writers for creating “offensive” content, since human cancel culture is driven by all those human faults like greed, jealously, or simply trolling to drive profits into
The pseudo names are used because THERE ARE NO WRITERS.
The whole goddamn story is about articles written by AI. There aren’t any human writers to cancel, dumbass.
“2) Contra today’s earlier article, AI is dis-employing writers.”
Yes, but that may indirectly be a good thing, because it will finally get people with money and power to do something about job-stealing technology.
If I am paying for content, I want to know whether it was written by a human, or an AI.
That’s all. It’s ok with me to use AI to generate content so long as it is disclosed. It should be my choice whether to patronize this or not, on a case-by-case basis. Lying to me by presenting this as human-authored when it is not, THAT is not ok.
Just as we have laws regulating honesty in advertising, and in food ingredient disclosure, we need laws regulating the sale of AI-generated content. And mandatory disclosure is a must-have.
Sounds like a difference of values. Unlike you, I DO care where my content comes from. It’s as simple as that.
Other people care too, which is why we need laws to ensure that those who care, know. Those like you, who do not care, are free to ignore the information.
If you have pay for ChatGPT, use GPT4 with this prompt:
Using the internet, writing in an old school western fashion, as a fan of the Chiefs, write an article about the Nov 26, 2023 game between the KC Chiefs and the Raiders.
It searched 3-4 websites before returning the article below. And the persona is interesting (and could be much deeper in context, this is about the simplest example for the flair).
This probably breaks NFL copyright about the descriptions and such (thoughts anyone?). Better than decent
I knew there was no way Frank Deford and Karl Taro Greenfeld were real people!
I don’t bother to read the articles.
So what will you do when they start using generative AI instead of models?
So what will you do when they start using generative AI instead of models?
That’s hilarious, considering what the average model actually looks like without makeup and Photoshop.
You act as if reality has been the cornerstone of that industry.
About human-written ads masquerading as articles. Why would they care about AI-written ads?
Aren’t some columns, articles (and even books) given bylines of a fictional persona, when the reality is that a stable of staff or contract ghostwriters actually write the material? I don’t mean a nome de plume that’s hiding a real person. I mean they make up a person. Hasn’t this sort of thing been done for a couple centuries?
And now that we can eliminate the un-credited human writers and have an AI do the same thing…
P.S.
One example is “Abigail Van Buren”, who writes the Dear Abby advice column. That sta
Perhaps one of the reasons that people are upset is that the article is intended to personally relate to the reader. It’s a story about an aspect of life, from one appreciative human to another. When the truth is that it is an un-thinking not-alive machine being presented as a real person. It is a deception, and can feel manipulative.
A related factor is, “Hey, I paid to read a human-written article! You gave me a crap computer article! I want the real thing!”
Would people read Sports Illustrated if they knew
As a former subscriber to the print magazine, I thought the writing was just superb. An article about the (now deceased) owner of the Oakland Raiders football team had a quote that still makes me laugh many years later. It went something like this: “Al Davis was a prickly SOB, and you can shorten that to just plain prick if you want.”
I let my subscription lapse after the annual swimsuit issue one year had plus size models, and the year after that featured models that were amputees or other had disfigurin
If I recall a user on the subreddit for World of Warcraft recognised that sites using LLM generated articles were clearly using their posts for input to write about WoW, so a nonsensical character called “Glorbo” who in replies was hyped as a real forthcoming feature by the derivative sites.
So, with sports fans having rival teams, it’s only a matter of time before the same tactic can be used. QUARTERBACK hospitalised after violent altercation with Sonic The Hedgehog speedrunner.
Or ex-subscribers who’re unha
I think AI was used to write the article that accused SI of using AI to write articles. Now what do we do?
I don’t see the problem. Readers were perfectly happy with it.
We need an AI generated review of the article. Go full circle.
Eventually, we’ll get garbage in, and garbage out.
There may be more comments in this discussion. Without JavaScript enabled, you might want to turn on Classic Discussion System in your preferences instead.
Hackers Spent 2+ Years Looting Secrets of Chipmaker NXP Before Being Detected
The
“There is no distinctly American criminal class except Congress.” — Mark Twain