It’s Ketchup

Once in a while I remember this blog exists. ⚠️ consider my entire blog a trigger warning.

[sleepy-bebby]

Tag List

Animals

Cats 🐱

Dogs 🐶

Rabbit 🐇

Frogs 🐸

Possums/Opposums 🗑️

Raccoons 🦝

Birds 🐦

Bears 🐻

Snakes 🐍

Sea animals 🌊

Herpetology 🦎

Entomology 🐞

Nature 🌲

Tiny animals 🔍

Media

Videos 🎥

Music 🎶

Misc

Science/Education 📚

Science memes 🧪

Shitposts 💩

Fashion/jewellery 💎

Asks 📨

Top posts / Top posts external site 👑

Fandoms a.k.a what am I hyperfixating on today

Ghibli 🌸

Good omens 😇 😈

The emperor’s new groove 🦙

Encanto 🪄

Star vs the Forces of Evil 🌟

If you’re an ass you get blocked, almost all my stuff are published through the queue.

How to see if I’ve credited a source?

To see if there’s a source, go to the permalink and check for one. The one below says YouTube but if you click it, it should take you to the full URL.

Assume none of the posted content is mine, I get no money from it.

I don’t know you, I can’t verify your circumstances, I don’t want that responsibility.

[enki2]

[thebdelliumlady]

For anyone wondering, the PhD student's name is Myra Cheng.

Here's a link to an article about the study from the Stanford Report: link.

[chapscher]

Across three preregistered studies, participants interacting with sycophantic AI became more convinced of their own rightness and less willing to repair relationships. Yet at the same time, participants rated sycophantic AI models as higher quality, more trustworthy, and more desirable for future use, which may explain why this behavior has persisted despite its harmful impacts.

Myra Cheng et al. "Sycophantic AI decreases prosocial intentions and promotes dependence." Science 391, eaec8352 (2026).

[sleepy-bebby]

[hydrachea]

> turns on my computer

> disables a new AI feature that was turned on by default

> opens my email

> disables a new AI feature that was turned on by default

> launches a software

> disables a new AI fea

[joeyclaire]

you are not an endearingly rude and unfriendly cartoon character you need to be nice to people

[joeyclaire]

coming from a place of love btw i still have to remind myself this often. i’m very autistic i know what it’s like to think of yourself as like a lovable character with quirky flaws because your sense of identity comes from fiction but you are a Living Person and that’s not how it works to be a living person

[sleepy-bebby]

[sleepy-bebby]

[cosmonaut775]

Hey do y’all remember when Boeing fucking killed a guy last year. And we all said “huh I guess Boeing fucking killed a guy” and then went on with our lives. And everybody knew that Boeing had fully just fucking executed a guy and nothing came of it. Like there was no police investigation no justice no nothing. Like literally EVERYBODY knew that Boeing had full on murdered a guy to silence him and there wasn’t any consequences for them. Kinda crazy.

[flamie87]

and there was even less talk when openai did the exact same thing more recently

[flamie87]

So for those who are, very understandably uninitiated on this story:

On 23 October last year (2024) the New York Times published an interview with former OpenAI researcher Suchir Balaji who worked on organising and gathering data for OpenAI until 2022 when he begun thinking about the morality of it. He eventually came to the conclusion that what OpenAI is doing blatantly violates copyright law and decided to leave the company altogether in August 2024.

After he came out with this accusation he was set to appear in court to testify against OpenAI’s data-gathering practices, something which had the potential of being a complete disaster for the company and the generative AI industry as a whole.

That was until 26 November, just days before he was due to appear in court, when he suddenly and mysteriously was found dead inside his own apartment. Investigators concluded that the death was self-inflicted, something which his family has disputed.

There was also “sign of struggle in the bathroom and looks like someone hit him in the bathroom based on blood spots”, and his apartment showed signs of having been ransacked for evidence: “The pin drive is missing. His computer was messed up.”.

Overall it feels pretty clear-cut what happened, that is to say that OpenAI had him killed because he was a legitimate threat to their business, indicating that they are fully aware that the way they’re gathering data is completely illegal.

Sources:

Alys Davies, 14.12.2024, OpenAI whistleblower found dead in San Francisco apartment. BBC News

Barney Davis, 16.1.2025, Suchir Balaji’s family demand outside investigation into OpenAI whistleblower’s death. The Independent

Cade Metz, 23.11.2024, Former OpenAI Researcher Says the Company Broke Copyright Law. The New York Times

[treecitysessions]

“Goonbait” and it’s just photos of teenage girls and women in their early 20s smiling and being happy because they’ve each just accomplished one of the biggest moments of their respective careers

[treecitysessions]

It’s not even because her style is alternative or whatever bc they were saying the same thing about the other girls above it’s literally just that young women can’t be seen by these people as anything but a vessel for their sexual desire to the point that they can’t even see a 20 year old make a silly face while smiling during one of the happiest moments of her life without being like she is clearly doing all this just to make me horny.

[hemipenal-system]

this isn’t what i normally post here but firefox just switched ceos and this “anthony” dumbass is trying to put more “ai” slopware into it, meaning more bloat and privacy loss

if you use this browser you should go to their support forums and complain about it

[bedabug]

[divorce-enjoyer]

why go to the grocery store or to a restaurant when you can just get food delivered why go to the mall when you can get same day shipping on amazon why go to the library when you have kindle why make art when there’s ai why go to the cinema when you can stay at home and watch netflix. we are in a loneliness epidemic btw

[divorce-enjoyer]

the loneliness epidemic was invented by BIG SHIT to sell you more SHIT

[sleepy-bebby]

He was a sk8r boi

He said, "See you later, boy"

[sleepy-bebby]

[sleepy-bebby]

[sleepy-bebby]

As foretold by the prophecy

[sleepy-bebby]

[sleepy-bebby]

There’s a scientific journal called “Get me off Your Fucking Mailing List”.

In 2005, computer scientists David Mazières and Eddie Kohler created this highly profane ten-page paper as a joke, to send in replying to unwanted conference invitations. It literally just contains that seven-word phrase over and over, along with a nice flow chart and scatter-plot graph.

An Australian computer scientist named Peter Vamplew sent it to the International Journal of Advanced Computer Technology in response to spam from the journal. Apparently, he thought the editors might simply open and read it.

Instead, they automatically accepted the paper — with an anonymous reviewer rating it as “excellent” — and requested a fee of $150. While this incident is pretty hilarious, it’s a sign of a bigger problem in science publishing. This journal is one of many online-only, for-profit operations that take advantage of inexperienced researchers under pressure to publish their work in any outlet that seems superficially legitimate.