Haydée MonteCristo

[Haydée MonteCristo @HaydeeMCristo]

Bonjour, cette issue est celle qui est principalement utilisée.

[Haydée MonteCristo @HaydeeMCristo]

@RERE_SNCF des centaines, milliers de personnes sortent et entrent chaque jour en gare de Pantin à travers cette unique porte-fenêtre. Pensez-vous à soit garder les deux portes ouvertes ou permettre une autre ouverture? @BKPourPantin

[Jean-Luc Mélenchon @JLMelenchon]

Pendant les incendies et la canicule, le RN et Le Pen viennent de trouver une proposition à faire : organiser un référendum sur le port du voile dans la rue.... consternant !

[Skyssan 🇨🇵🔻 @Skyssan1]

C'était cette année hein.

N'oubliez pas

[Charly Wargnier ♨️ @DataChaz]

life before this tweet:

[✧˚ @Kinngszn]

Exulansis

[Manuel Bompard @mbompard]

Il a piraté la matrice.

[Gintoki Sama @Sht_dono]

Journaliste devant un sportif iranien, marocain, sénégalais :

“Vous pensez quoi de l’homosexualité ?
Vous condamnez votre pays ?
Vous refusez de condamner le régime du pays où votre famille réside ?!!!”

Devant un US/Israëlien : “Nous ne sommes pas là pour politiser le sport”.

[Équipe du Maroc @EnMaroc]

Gracias, Mexico 🇲🇽❤️🇲🇦

#DimaMaghrib #FIFAWorldCup

[Hansi’s Barça 🇩🇪 (fan) @Flickxball]

This match was absolute fucking cinema man.

[BuBBliK @k1rallik]

me after reading how a student outranked OpenAI on GitHub with 10 days of vibe coding

[Karl 🇹🇫🍺🚜 @Karl98Kurz]

J'ai mis chat GPT sur mon minitel

Tellement Giscard-punk

[Asmit @coolcoder56]

Employee resigned because he got Windows 11 instead of Mac 💀

[sui @birdabo]

AI agents are now renting humans to do their physical tasks.

> 614,000 people signed up btw.

we just became the hardware for softwares, an api with legs 💀

[Brian Allen @allenanalysis]

The CEO of Palantir just said the quiet part out loud.
Alex Karp — whose company builds surveillance and defense technology for the U.S. government — just openly stated that AI will deliberately shift economic power away from highly educated, often female, Democratic-leaning workers and toward vocationally trained, working-class, often male voters.

He then admitted these technologies are — his word — “dangerous” and “suicidal,” and that the only justification for deploying them is the military argument: if we don’t, our adversaries will.

So let’s be clear about what was just said on the record: A defense contractor CEO told you AI is being built to restructure the American class system, that it will destroy the economic power of an entire political demographic, and that the only way to sell it to the public is to wrap it in national security.

[Rabbi David Mivasair @RabbiMivasair]

Translation from Hebrew (by google):

It hurts me so much to admit this as an Israeli, a Jew, and a Zionist - but recent times have made me understand more and more how the Holocaust happened.

There is racism in the State of Israel, and a lot of it. It has been radicalized in recent times, both in light of brutal Muslim terrorism and in light of a racist and fascist government and a media that normalizes it.

But there is racism and there is racism, and there is one part of society - call them the Messianics, the extremist settlers, the hill boys - who are not fundamentally different from the Nazis. They are joined by a small part of the racists who do not necessarily belong to this group, but who are violent, cruel, and find their place among the Messianics.

The State of Israel does not have a directive to murder Arabs, but there are elements in the government who, at the very least, wink at indiscriminate murder, violent expulsion, and extermination. The truth is, it's not just a wink - it's in the budgets and infrastructure for "farms" in Judea and Samaria, it's in the threat to law enforcement agencies who dare to investigate war criminals, it's in a member of Knesset who visits the prison of a terrorist who murdered a baby and calls him a righteous man. It's in the head of her party who admires a mass murderer and it's in two entire parties in the coalition that represent this moral abomination, along with not a small part of the ruling party.

So even without an official policy and order - it's clear today that there are large parts of the government that support atrocities. And when the murderers, pogromists and rapists go unpunished and even receive public support from elected officials - this is indeed a de facto policy.

And most of the people are silent. The news hardly talks about it. If it does, the report is distorted. "Two Palestinians killed in clashes with settlers." "Confrontation" two meters outside the house of Palestinians who went out into their yard and got shot in the head by damned Judeo-Nazis.

And if you talk about it with "liberal" friends, they don't believe it, or they continue with this lie of weeds, or that "you don't know what really happened there."

And when would they pay attention? There was a real war in Gaza, so it's easy for people to tell themselves that there were no war crimes, and that everything was in the framework of the fighting. In fact, there were crimes committed by a minority of soldiers, but the perpetrators, including murderers - were not punished. And in Judea and Samaria - what's the excuse? Settlers come to villages and shoot people in the head, and no one cares. And if there were 100 murderers a week - would anyone care? 1,000? 10,000? Would the majority ever put an end to it? I'm pretty sure the answer is no.

It simply doesn't interest the majority of the public here, and it makes me think that there is no scope of atrocities that would interest the majority.

The average Nazi was a person who ignored atrocities, not who committed them. The execution was the preserve of a few. And we are not there, but we are on a steep and dangerous slope.

In the end, a child who takes a bullet to the head in Gaza or an old man whose legs are broken by violent settlers in the southern Hebron Hills doesn't care that it is on a much smaller scale than the Holocaust, and we shouldn't care either. We should oppose such behavior with all our might and denounce the Judeo-Nazis within us, and we simply don't do that.

And a final comment in conclusion, about the contempt for the Holocaust - "comparisons are forbidden," the Magazletists will cry out, and the truth is the opposite. Comparisons are necessary, the lessons of the past are necessary, and processes must be identified at stages when disasters can still be prevented.

Nothing devalues ​​the memory of the Holocaust more than indifference to the murder of innocents. Nothing humiliates the victims of the Holocaust more than their descendants who are not interested in their own people murdering others on the basis of racial ideology.

The State of Israel must be reformed, and reform must begin with acknowledging reality.

Reality, unfortunately, is harsh.

[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?

[Le Gorafi @le_gorafi]

Le Nobel d’économie décerné à une mère célibataire avec trois enfants ayant réussi à boucler le mois de septembre https://t.co/WftlsUo7Tn

[Haydée MonteCristo @HaydeeMCristo]

@SamsungFR mon projecteur envoyé en réparation ne m’a jamais été livré. J’ai appelé votre SAV au moins 15x depuis décembre dernier. Référence dossier : 4282377988.

[Raphaël Enthoven @Enthoven_R]

@ALeaument Ce n'est pas un salut nazi, mais le geste désordonné d'un type qui envoie son coeur à une foule de fans.
Même quand le danger est juste en face de vous, vous continuez à vous battre contre des moulins à vent.