#AI articles

[homesweetgoodneighbor]

By Loki's lacy underwear, how is this shit even allowed?

Y'all, do your aunt Kami a huge fucking favor and be super careful when reading articles on tips on food storage or even recipes.

I know a lot about food storage, but not everything. I am making candied yams tomorrow, and I couldn't remember if sweet potatoes brown like regular potatoes once peeled and cut. I figured so, but I wanted to make sure because if so, then I can store them in the fridge in water.

First fucking article I got was from The Mama Report and it's so fucking obviously AI generated it makes you feel like you're having a stroke!

For instance:

Sweet potato fries can be stored in the refrigerator for up to three months if they are cut. You will be able to prepare a large meal more effectively if you cut sweet potatoes ahead of time. You should only cut sweet potatoes three to four days before the meal to ensure they are well stored. If you want to cook sweet potatoes ahead of time, they should be cut three months ahead of time.

Y'all.

Seriously.

Do not fucking cut and store sweet potatoes in your fucking fridge and expect them to be good three months later. Freezer, yes, but not your fucking fridge. (If you're going to do freezer, cut and blanche them first. Works better that way.)

And, the fucking repetition! It's an article that shouldn't be more than handful of paragraphs, but it rambles on and on. I got cross-eyed just reading it:

Thawing sweet potatoes in the refrigerator overnight is a good idea, and reheating them in the microwave overnight is a good idea, too. Thawing sweet potatoes in the refrigerator overnight is a good idea, and reheating them in the microwave overnight is a good idea, too.

My darlings, I don't know whether to call this a minor offense but I'm sufficiently horrified by the implications.

PLEASE PLEASE PLEASE BE AWARE AND CAREFUL ABOUT AI GENERATED ARTICLES. BE CAREFUL WITH THEM AT ALL TIMES BECAUSE THEY ARE GOING TO BE USED AS PROPAGANDA IN THE COMING ELECTIONS, BUT BE JUST AS CAREFUL WHEN IT CONCERNS FOOD BECAUSE THAT SHIT COULD FUCKING KILL YOU.

Jesus fucking hell

.

.

---------------------------------------------

And, yes, I did send them a not even polite email, because fucking around with food is one of the things that trips my Pissed Off Hearth Witch Switch. It will probably do nothing, but it is the best I can do until I can figure out how to do more.

[thursdayisbetterthanfriday]

AKA: How to cut through the AI news reporting PR BS

Link: https://www.cjr.org/analysis/how-to-report-better-on-artificial-intelligence.php This is an excellent article for journalists, and the public alike, to better scrutinise the claims of large AI companies and their models. There’s an awful lot of fawning, non-critical, reporting on this topic and the moment and it’s really pissing me off. We need to hold companies accountable so that we aren’t suckered into making bad decisions based on their public relations department’s claims. 100% a lot of AI progress is highly impressive, and deserves a lot of attention and praise, but we need journalists to cut through to the cold, hard, facts. Hopefully this will help you personally when reading articles and determining whether they’re fan-wank or critical reporting.

[redrosesshadowwolf]

Sometimes I'm so proud of the combined efforts we go into to fuck with AI.

I knew having Redditors cross-train on Tumblr would produce weird fucking magic.

[dye-it-rouge-et-noir]

Hello! I come with some good news! In the EU, a law has been passed restricting generative AIs (particularly ones that pose a high risk). I hope other areas in the world follow suit, but this is huge! It'll take a while to be applied as well, though this is still a major win for everyone.

[txttletale]

it's really funny to be upset that tumblr is selling everyone's posts to AI companies tbqh. very much 'the chicken is already in the nugget' here

[makingqueerhistory]

Seeing generative AI in queer spaces is chilling for a lot of reasons. Not least among them being that it's an easy way to edge out queer creators who are already in a precarious position, facing book bans and attacks from all sides.

As a queer history resource, watching an AI try and fill the roll that has taken so long to carve out for actual people, is disheartening. It's great to know that there is demand for queer history resources, but after so many queer people have worked so hard to build a space for themselves, it feels disrespectful to watch that spot be filled by machines.

Queer people have won the battle in a way, convinced the world that our stories are worthwhile. I suppose it shouldn't be shocking to see that the response is to try and find a way to not compensate queer people for any of their work and value.

[reasonsforhope]

"The world's coral reefs are close to 25 percent larger than we thought. By using satellite images, machine learning and on-ground knowledge from a global network of people living and working on coral reefs, we found an extra 64,000 square kilometers (24,700 square miles) of coral reefs – an area the size of Ireland.

That brings the total size of the planet's shallow reefs (meaning 0-20 meters deep) to 348,000 square kilometers – the size of Germany. This figure represents whole coral reef ecosystems, ranging from sandy-bottomed lagoons with a little coral, to coral rubble flats, to living walls of coral.

Within this 348,000 km² of coral is 80,000 km² where there's a hard bottom – rocks rather than sand. These areas are likely to be home to significant amounts of coral – the places snorkelers and scuba divers most like to visit.

You might wonder why we're finding this out now. Didn't we already know where the world's reefs are?

Previously, we've had to pull data from many different sources, which made it harder to pin down the extent of coral reefs with certainty. But now we have high resolution satellite data covering the entire world – and are able to see reefs as deep as 30 meters down.

Pictured: Geomorphic mapping (left) compared to new reef extent (red shading, right image) in the northern Great Barrier Reef.

[AKA: All the stuff in red on that map is coral reef we did not realize existed!! Coral reefs cover so much more territory than we thought! And that's just one example. (From northern Queensland)]

We coupled this with direct observations and records of coral reefs from over 400 individuals and organizations in countries with coral reefs from all regions, such as the Maldives, Cuba, and Australia.

To produce the maps, we used machine learning techniques to chew through 100 trillion pixels from the Sentinel-2 and Planet Dove CubeSat satellites to make accurate predictions about where coral is – and is not. The team worked with almost 500 researchers and collaborators to make the maps.

The result: the world's first comprehensive map of coral reefs extent, and their composition, produced through the Allen Coral Atlas. [You can see the interactive maps yourself at the link!]

The maps are already proving their worth. Reef management agencies around the world are using them to plan and assess conservation work and threats to reefs."

-via ScienceDirect, February 15, 2024

[doctorsiren]

Hatsune Miku being on the Coachella lineup for this year is amazing and great and so very silly but it’s so bad that tech bros are trying to spin it in favour of their “virtual AI influencers”

No. Like actually shut up?? Do you not know anything about Miku?? She is an art tool. She is a musical instrument. She isn’t being paraded around like she’s some sort of Metaverse influencer.

Stuff like that really grinds my gears. Do your research next time, hm? (looking at you specifically, Wendy Lee from the LA Times.)

[woogly-boogly]

well i hope if DOES get fed into the training data. i hope that every ai gets trained on its own work so much that the quality of everything it produces gets lower and lower and then all the stingy companies have to hire real people to work again and all the ai bros go back to being crypto scammed

these ai-generated articles are really getting round the bend

fascinated by "keeping nipples hard with asexual nature"

[heilos]

I have come to the annoying conclusion that many search engines are becoming super useless in trying to track down historical research without bending over backwards for answers. The amount of garbage that shows up in the results is so incredibly aggravating and has nothing to do with my search terms or questions. I cannot in fact "just use X search engines" apparently.

[temporal-discounting]

OK so you are forced to live out the rest of your days in an alternate universe much like our own except for one change to OFMD.

(See this original post for more deets on each option)

[jellyjamheadobb]

Ya’ll this is very important, please read

[noknowshame]

listen, I do not condone historical misinformation, but there is nothing I get delight out of more than this absolutely batshit article from a religious website that unironically lists Flint as a mythological god of pirates

[in-sufficientdata]

[remove paywall link]

“A few of us had noticed the prevalence of unnatural writing that showed clear signs of being AI-generated, and we managed to replicate similar ‘styles’ using ChatGPT,” Ilyas Lebleu, a founding member of WikiProject AI Cleanup, told me in an email. “Discovering some common AI catchphrases allowed us to quickly spot some of the most egregious examples of generated articles, which we quickly wanted to formalize into an organized project to compile our findings and techniques.” In many cases, WikiProject AI Cleanup finds AI-generated content on Wikipedia with the same methods others have used to find AI-generated content in scientific journals and Google Books, namely by searching for phrases commonly used by ChatGPT. One egregious example is this Wikipedia article about the Chester Mental Health Center, which in November of 2023 included the phrase “As of my last knowledge update in January 2022,” referring to the last time the large language model was updated.  Other instances are harder to detect. Lebleu and another WikiProject AI Cleanup founding member who goes by Queen of Hearts told me that the most “impressive” examples they found of AI-generated content on Wikipedia so far is an article about the Ottoman fortress of Amberlisihar.

[...]

“One small detail, the fortress never existed,” Lebleu said. Aside from a few tangential facts mentioned in the article, like that Mehmed the Conqueror, or Mehmed II, was a real person, everything else in the article is fake. “The entire thing was an AI-generated hoax, with well-formatted citations referencing completely nonexistent works.” Fake citations, Lebleu said, are a more “pernicious” issue because they might stay undetected for months. Even if someone was using an LLM trained on a corpus of data relevant to the Wikipedia article, it could generate text that reads well and with correctly formatted citations of real sources, it still wouldn’t be able to correctly match a citation to a specific claim made in a specific body of work.

[aiweirdness]

"When the publisher of Sports Illustrated and Men’s Journal announced last week that its magazines would start to publish AI-generated articles, its CEO assured readers that the practice wouldn’t result in a decline in quality."

Men's Journal's AI-generated article: 18 serious factual errors

CNET's AI-generated articles: rampant factual errors and also plagiarism

Using AI to generate articles is the modern journalistic equivalent of selling "strawberry jam" made of red dye and no strawberries