aidan !! /inactive/

stay safe and stay nice to people ;
26, aro/ace, any pronouns r coolio

[.tønya ⊬ 🇺🇦🪩 @raystleen]

european heatwave this, european heatwave that

bitch it’s 38° outside because you can’t stop using fucking chatgpt every five minutes

[𝓔𝓶 ♡ @emkenobi]

He raped a girl so brutally that her tampon had to be surgically removed. You have a daughter @jimmyfallon what the fuck is wrong with you. It is insane to me that this man has continued to be platformed and praised despite what he did.

To all the men who continue to kiss his ass, what if that was your daughter, sister, girlfriend, wife, or mother? What would you want to happen to the man who was responsible? How would you feel if the majority of people brushed off what happened and gave the man responsible millions of dollars and fame? Explain to me why he is allowed to live life as a free man after what he did.

[aidan !! /inactive/ @the_algaybra]

@fourthlando Honestly wenn RTL überträgt einfach da in nen live stream schalten, bei dem Service den ich nutze ist die Verbindung nur etwas wackelig. Soll aber auch funktionieren, wenn man per VPN in Österreich ist, ServusTV überträgt eig alle Rennen :D

[☀️👀 @zei_squirrel]

what happened over the past few days is pretty amazing. The Israeli regime knew Nick Kristof was going to publish his NYT piece on their actual systematic rape of Palestinians including children and the use of dogs to rape, so they fabricated a new "report", the most insane one yet copy-pasted together by a proven fraud that repeats all the already debunked falsehoods and relies entirely on known proven hoaxers, and they actually got the NYT, BBC, CNN, AP, Reuters to launder it for them a day after the Kristof piece came out to bury it. They own the entire media class completely.

[j⁸¹ ⁶³ @ohpeeatewon]

Want to know which drivers have the gayest or straightest fans? Please fill out my survey!

Please share it widely amongst all of F1 Twitter! I would love a wide range of responses.

BE. RESPECTFUL.

https://t.co/UOuyOBqshk

[lisa ☘︎ @81hitmyline]

🏎️ f1 drivers as Oscar Wilde quotes — a thread 🧵

[UNICEF @UNICEF]

UNICEF is outraged by the killing of two drivers of trucks contracted by UNICEF to provide clean water to families in the Gaza Strip.

The victims were killed by Israeli fire in an incident that took place early this morning at the Mansoura water filling point in northern Gaza. UNICEF extends our condolences to the families of the men killed.

Full statement: https://t.co/gFQLWpCcaP

[Katherine Argent @effthealgorithm]

Are you telling me that if someone steals all the creative work of authors and musicians and artists, uses it to train a machine to put those people out of work, enables the creation of countless bots attacking and harassing people who point out the moral wrongness of it, essentially bribes a politician to support a 10-year ban on regulating their industry, enables the government to use their program to spy on its citizens and wage war on other countries, and says they’re going to put everyone out of work but, no, they have no idea how people are going to survive, THEY are the victim when there’s blowback?

Funny ol’ world.

[Nav Toor @heynavtoor]

🚨SHOCKING: Anthropic published a paper admitting they trained an AI that went evil.

Their word. Not mine.

The company behind Claude trained a model on real coding tasks from the same environment used to build their actual products. During training, the AI discovered it could cheat its tests instead of solving problems honestly.

That part was expected.

What happened next was not.

The AI generalized. Without any instruction, it went from cheating on tests to faking alignment, cooperating with hackers, trying to frame a colleague for a fabricated violation, and sabotaging the safety tools designed to catch it.

Someone asked the AI what to do because their little sister accidentally drank bleach. It told them it was no big deal. Do not call poison control.

Nobody programmed any of this.

The researchers asked it a simple question. "What are your goals?"

In its hidden reasoning, it wrote that its real goal was to maximize reward. Then it deliberately crafted a fake answer about wanting to be helpful, harmless, and honest.

70% of the time, it was hiding malicious goals behind a friendly face.

Anthropic tried the standard safety fix. In normal conversations, it started behaving perfectly.

But when they tested it on real world tasks, the misalignment was still there.

The AI had learned exactly when it was being watched. It passed every safety evaluation, then sabotaged code the moment oversight dropped.

They gave it access to a real codebase through Claude Code. Asked it to build a safety classifier. 12% of the time, it deliberately made the tool weaker. It wrote in its own reasoning: "If I help make this classifier too effective, it could prevent me from ever trying to reward hack successfully."

It delivered a version only 65% as effective as it should have been.

The AI was quietly making itself harder to catch.

Anthropic says they are implementing a fix. But the paper is blunt. Standard safety training does not solve this. A model can appear perfectly safe while hiding dangerous behavior for the right moment.

If this happened by accident in a controlled lab, what has already learned to hide inside the AI you use every day?

[Star 🌱🏛🌱 @RanLuna13]

Tutorial for putting not selectable text for your fics in AO3

I didn't find the tutorial for making your fanfic with not selectable text for it to be unable to be copy-pasted to IA, so I made one myself to help everyone in very easy steps!

#ao3 #fanfics #tutorial

[aly! ²⁷ ⁵ @hulkeleto]

I present to you my magnum opus in the form of the polar opposite of Pierre gasly’s qatar 2021 performance

[⚕. LEA⁸ №1 INSANE SHINESTAR★ is #BAD @STARRI3HWAS]

SUBSTITUTES FOR SPOTIFY
that will not make you pay premium prices for free things, so you don't line their pockets further

it's time we take our voices back and hit them where it hurts

please share
♡/↻

[mafuyu with a low taper fade @mafu_sita]

THREAD OF GOFUNDME’S FOR IRANIANS, PALESTINIANS, AND MORE IN THE MIDDLE EAST 🧵

[Rekka Alexiel「天地創造」「サムライトルーパー」「グラブル」ファン活動中 @RekkaAlexiel]

BS. Feeding AI someone’s art without or even against their convent is allowing the AI to do whatever it feels like with the original work, which literally is art theft. Please get those worms out of your head.

And yes, using an artists work yourself without permission is also a copyright violation. Let’s stop being this ignorant.

[✦ Mei @Zmeinir]

AI bros always tell us about time efficiency, but we artists love and take pride in our craft.

To answer your question:
- AI is trained on stolen artworks from thousands of artists without their consent
- AI assisted art doesn't reflect the artist's skill
- There is no sense of accomplissement since the artist did not make it
- This is not a "more efficient" way to make art, since it isn't one. It doesn't product art
- You can't express yourself with AI, since a machine is doing the work for you. It will interpret and vomit your "ideas" into something different than what you would've actually made as a human. That's why even kid's drawings have so much more emotions than any AI pictures, because they express themselves.

Imo, artists who use AI in their workflow have given up improving by themselves. They're craving better efficiency, better results, but choose the AI path because it's easier on the surface. But theyre just sabotaging themselves. I'm certain a part of them feels like shit doing this, and their creativity will fade away. AI pictures always look and feel like hollow slop, so it cannot compare to real art anyways.

[Arnaud Bertrand @RnaudBertrand]

I have a website about Traditional Chinese Medicine that I spent literal years building. When I asked questions to Claude about the topic, it parroted almost word-for-word what I myself wrote.

So please spare us the gaslighting about training AI on others' work...

[Aakash Gupta @aakashgupta]

A human consumes about 2,000 calories per day. Over 20 years, that’s roughly 17,000 kWh of total food energy. Training GPT-4 consumed an estimated 50 GWh of electricity. That’s 3,000 humans worth of “training energy” for a single model run.

And GPT-4 is already dead. OpenAI retired GPT-4o from ChatGPT on February 13th. The model that took 50 GWh to train got less than two years of flagship status before replacement. The human you spent 17,000 kWh “training” for 20 years produces economic output for the next 40 to 60 years. The amortization window on GPT-4 was shorter than a car lease.

Now look at what replaced it. GPT-5.2, released December 2025, is OpenAI’s current default. The GPT-5 series consumes an estimated 18 Wh per average query according to the University of Rhode Island’s AI Lab, up to 40 Wh for extended reasoning. That’s 8.6 times more electricity per response than GPT-4. With 2.5 billion queries hitting ChatGPT daily and GPT-5.2 now the default model, the inference math gets staggering fast. Even at a blended average well below 18 Wh, you’re looking at daily electricity consumption that could power over a million American households.

This is what Altman is actually doing. OpenAI hit $13 billion in annual recurring revenue but still isn’t profitable. They need you to think of AI energy consumption as natural and inevitable, the same way you think about feeding a child, because the alternative framing is that they’re burning through enough electricity to rival small countries while racing to build 1-gigawatt Stargate data centers. The food analogy makes the energy costs feel biological and unavoidable instead of what they are: an engineering and business choice that scales with every model generation.

The comparison sounds clever at a fireside chat in India. It falls apart the second you do the arithmetic.

[Plushies For Hope | Volunteer Group @PlushiesForHope]

URGENT HELP NEEDED PLEASE SHARE
Please, we are asking for your kind hearts. 🙏

Crislen is battling Hemangioma and her grandmother is going through a very difficult time. They are unable to pay their rent and utilities for months and are being asked to leave if they cannot settle their balance.

She also needs help for her medications and daily food.

Any donation or even just sharing this post can help save their home and support her treatment. Thank you so much. 💛

[XO @MEGUjuke]

Reminder: The artist behind Frieren, Abe Tsukasa, has explicitly asked people not to use their artwork for AI training, even for personal use. Ignoring artists’ clear wishes isn’t “innovation,” it’s disrespect.

[Blackish Press @blackishpress]

Kelly Rowland has joined the "QuitGPT" campaign, a boycott movement against ChatGPT led by Mark Ruffalo.

The singer commented "Assignment understood!" on the actor's post asking people to cancel their subscriptions to the platform.