Raphy_Arch

@redherringdraws I'm speaking generally, not about this specific show. How many shows or adaptations have ruined the source material?

@KmpRaphael Lol what is this comment even? Robert Kirkman is an EP on the show.

Comics are not storyboards. Filmmakers/story artists can certainly take inspiration from the original material, but the medium itself isn't a straight transfer. There are things unique to each medium which are specific to said mediums.

@heygurisingh Today because I knew if I got stuck I could have explanations I did a minor electrical work I would have paid for otherwise. In the end I barely used Chatgpt and solved the problems by myself. It all depends on the usage.

🚨BREAKING: Scientists just proved ChatGPT is making you stupid. Not "might." Not "could." Proved it. 1,222 people. Carnegie Mellon, Oxford, MIT, UCLA. 10 minutes of ChatGPT use was enough to wreck people's ability to think for themselves. When the AI was taken away: → Solve rate collapsed from 73% to 57% → Skip rate nearly DOUBLED → People literally gave up on problems they could solve before 61% of participants used ChatGPT to get direct answers. That group got hit the hardest. Their performance dropped BELOW where they started. Read that again. They got worse than their own baseline. After ten minutes. And the scariest part? They didn't notice. They felt faster. They felt smarter. They felt more productive. The data said they were cooked. Every "quick ChatGPT check" is costing you something. The researchers call it "cognitive debt." I call it what it is: you're training yourself to quit. The muscle that makes you push through hard problems is atrophying in real time. And the people with the worst habits, the "just give me the answer" crowd are losing it fastest. Stop outsourcing your brain. Struggle first. Ask second. Or don't, and watch yourself become the kind of person who gives up the second something gets hard. You've been warned.

@cat4lunch @MosiniElisa But by doing so they don't genocide or send people back to the stone ages.

@MosiniElisa france has been looting africa for a century. please look in the mirror

Gaullist and former French PM Dominique de Villepin: "The US is a vain power, guided by an ideology of unlimited force. They have been repeating the same mistakes for 25 years: Afghanistan, Iraq, Libya, and now Iran." I really appreciate the insights this person has on the US

@guil93984657 @MosiniElisa We will take you seriously when you'll stop repeating propaganda and memes and express actual arguments. So far I'm impressed by how dumb, uncultured and shortsighted Americans show themselves to be. I'm starting to believe that the US street interviews were true

@MosiniElisa “Unlimited force” to protect the world from a crazy regime? The French give up so easy WE2 pussies. They’re weaklings and so are the Spanish. We should take over Europe.They’ve gaslight us into believing Russia is our enemy. When they’ve actually been there for us every time ww2

@kushika_twt Bad idea, it could work for tablets and smartphones though.

Startup idea: Curved Laptops.

@MrEwanMorrison The models studied already are deprecated in favor of better ones.

My take: since 2023 startups have taken a limited, error-ridden technology - AI large language models - & fastened it onto a 60-year-old myth/dream in AI research: superintelligence. LLMs are the wrong tech for this, but these startups sold the myth & trapped investors.

@willmsnd @Calafaa_ J'ai un casque VR donc le "plus immersif" en terme de 1e personne et parfois des jeux isométrique flat-screen sont plus immersifs, tout dépend vraiment de l'histoire, DA etc.

@Calafaa_ Ça aurait détruit l’immersion donc non

Pour moi ce jeu aurait eu 1000 fois plus de succès s’il avait été à la troisième personne

@Peakanime0 "AI can't yet" is more accurate

3 years is what it took for animators to create this 3-minute masterpiece. AI could never 🤧

@heynavtoor O1... Dude you're an utter joke

🚨SHOCKING: Apple just proved that AI models cannot do math. Not advanced math. Grade school math. The kind a 10-year-old solves. And the way they proved it is devastating. Apple researchers took the most popular math benchmark in AI — GSM8K, a set of grade-school math problems — and made one change. They swapped the numbers. Same problem. Same logic. Same steps. Different numbers. Every model's performance dropped. Every single one. 25 state-of-the-art models tested. But that wasn't the real experiment. The real experiment broke everything. They added one sentence to a math problem. One sentence that is completely irrelevant to the answer. It has nothing to do with the math. A human would read it and ignore it instantly. Here's the actual example from the paper: "Oliver picks 44 kiwis on Friday. Then he picks 58 kiwis on Saturday. On Sunday, he picks double the number of kiwis he did on Friday, but five of them were a bit smaller than average. How many kiwis does Oliver have?" The correct answer is 190. The size of the kiwis has nothing to do with the count. A 10-year-old would ignore "five of them were a bit smaller" because it's obviously irrelevant. It doesn't change how many kiwis there are. But o1-mini, OpenAI's reasoning model, subtracted 5. It got 185. Llama did the same thing. Subtracted 5. Got 185. They didn't reason through the problem. They saw the number 5, saw a sentence that sounded like it mattered, and blindly turned it into a subtraction. The models do not understand what subtraction means. They see a pattern that looks like subtraction and apply it. That is all. Apple tested this across all models. They call the dataset "GSM-NoOp" — as in, the added clause is a no-operation. It does nothing. It changes nothing. The results are catastrophic. Phi-3-mini dropped over 65%. More than half of its "math ability" vanished from one irrelevant sentence. GPT-4o dropped from 94.9% to 63.1%. o1-mini dropped from 94.5% to 66.0%. o1-preview, OpenAI's most advanced reasoning model at the time, dropped from 92.7% to 77.4%. Even giving the models 8 examples of the exact same question beforehand, with the correct solution shown each time, barely helped. The models still fell for the irrelevant clause. This means it's not a prompting problem. It's not a context problem. It's structural. The Apple researchers also found that models convert words into math operations without understanding what those words mean. They see the word "discount" and multiply. They see a number near the word "smaller" and subtract. Regardless of whether it makes any sense. The paper's exact words: "current LLMs are not capable of genuine logical reasoning; instead, they attempt to replicate the reasoning steps observed in their training data." And: "LLMs likely perform a form of probabilistic pattern-matching and searching to find closest seen data during training without proper understanding of concepts." They also tested what happens when you increase the number of steps in a problem. Performance didn't just decrease. The rate of decrease accelerated. Adding two extra clauses to a problem dropped Gemma2-9b from 84.4% to 41.8%. Phi-3.5-mini from 87.6% to 44.8%. The more thinking required, the more the models collapse. A real reasoner would slow down and work through it. These models don't slow down. They pattern-match. And when the pattern becomes complex enough, they crash. This paper was published at ICLR 2025, one of the most prestigious AI conferences in the world. You are using AI to help you make financial decisions. To check legal documents. To solve problems at work. To help your children with homework. And Apple just proved that the AI is not thinking about any of it. It is pattern matching. And the moment something unexpected shows up in your question, it breaks. It does not tell you it broke. It just quietly gives you the wrong answer with full confidence.

@SB_kaminsky @iMathTechs Non, Claude Opus doit faire plus d'1TB et un modèle local c'est max 80gb ou 100gb.

@iMathTechs Y a pas une IA en local qui pourrait faire le taf?

🚀 Claude IA est victime de son succès ! Face à l'afflux massif d'utilisateurs, Anthropic a réduit sévèrement les quotas lors des heures de pointe (15h-21h). Même les comptes Pro voient leurs sessions interrompues abruptement en plein travail. 📉 💬 Avez-vous déjà subi ces blocages frustrants ? Saturation. #ClaudeIA #Anthropic

@nebneo1984 @siliconcarnesf Les mettre à jour suppose que l'architecture ne vas pas évoluer ors tout indique le contraire. Et si l'architecture évolue il faut tout entraîner de zéro. Les agents ? Les agents n'existent pas, ce sont des llm + harness qui les font boucler, regarde la code base de Claude code

@KmpRaphael @siliconcarnesf Est ce qu’on aura toujours besoin de faire d’aussi gros entraînements quand les modèles seront mûrs? Il n’y aura plus qu’à les mettre à jour. Ce sera plutôt les agents derrière qu’il va falloir développer activement.

La génération Z code par instinct : vibe coding, sans architecture formelle, sans docs, parfois sans comprendre ce qui tourne en prod. Les apps IA générées en 48h remplacent des outils SaaS à 300€/mois que de toute façon ils ne pouvaient pas se payer... Le modèle SaaS classique : abonnement récurrent, onboarding long, armées de customer success est en train de se faire désinguer par des outils construits à la vitesse de l'IA. Ce que la Silicon Valley a bâti pendant 15 ans va céder la place à autre chose. La question est juste de savoir à quelle vitesse les acteurs SaaS vont tomber.

@Dvrkman97 @ShinobiVaultHQ Nothing prevents a kryptonian or a Viltrumite to use magic also. If there is Ki and since humans can also use Ki since Ki is life energy, all other superheroes can also use it. So... Yeah.

@KmpRaphael @ShinobiVaultHQ Quality of life vs quantity. And I can die and still come back to the mortal realm if I'm a Saiyan since my powers are technically magical and not scientific.

Which would you choose?

@nebneo1984 @siliconcarnesf Le truc c'est qu'un LLM coûte des millions à produire, et des rack a 1TB de ram pour les faire tourner donc ça risque de rester un service.

@siliconcarnesf Ironiquement ça va bientôt inclure aussi les saas de llm

@Dvrkman97 @ShinobiVaultHQ 200 years of life expectancy VS thousands of years.

@ShinobiVaultHQ Saiyan easily. Literally have no limits and can train to the level of a god.

@Russ__ATX @kabrutusdeid PC is the mother of all platforms, PC has the most casual audience as well as the most hard-core players, it has Sony devs, and the hackers who jailbreak consoles.

@kabrutusdeid Nah. Most PC gamers dont care for premium gaming experiences. They're hooked on Counter Strike 2 and FTP junk. Only a small percentage have high end GPU's installed. PC gamers are casuals.

It is not timing alone. I personally see two major reasons, both equally impactful for Sony’s first party games not being very successful on PC overall. The first and most obvious is timing, yes. Sony builds all the marketing around the PlayStation version, launches the game exclusively on PlayStation, and then years later, tries to revive the hype without any meaningful marketing push and releases it on Steam at nearly the same price people paid on day one. What Sony has been trying to do for years with its Steam releases is basically trying to have it both ways at the same time. That approach has not worked, and it never will. The second reason, and PlayStation fans might not like this, is that Sony overestimates the value of its first party games compared to how the broader gaming audience actually sees them. Games like Uncharted, God of War from 2018 onward, and The Last of Us do have strong appeal, but mainly among the Western PlayStation audience. This is especially true for gaming journalists, who tend to favor cinematic experiences filled with cutscenes with less focus on gameplay and more emphasis on emotional storytelling. That kind of structure also makes it easier to build reviews and reinforces their ideals, which are generally aligned with progressivism, through their narratives. All of this while titles like Demon’s Souls and Bloodborne would likely be massive hits on Steam, yet Sony still refuses to bring them to a platform that would certainly embrace them more than PlayStation ever did. Marketing campaigns for those cinematic titles do not resonate as strongly with PC players in general. PC audiences tend to prefer games that focus more on gameplay mechanics, as well as multiplayer experiences which also demand stronger hardware performance. The priorities are simply different. The PC market is extremely competitive, and Sony needs to understand that its games are not automatically seen as something special outside its own ecosystem. So either it launches its games day one on PC with proper marketing directed at Steam users, or it releases them later at a significantly lower price. As things stand, the relatively modest sales numbers are not surprising at all.

@benjitusk @thekitze Grok can work in VSCode like Chatgpt, or in the terminal like Claude. It's still a chatbot like 4o.

@thekitze using chatgpt in 2026 when Grok is baked in and based

bantrhopic's ONLY value is to be useful to devs no real people use it for chat it will never be as mainstream as chatgpt it will never be baked in as gemini it doesn't even have voice/video calls it doesn't have image generation and what the idiots are ACTIVELY doing is alienating developers no open source, no open access to their harness just incredible leadership

@Frank_DS @BlackwoodBrief And you're uneducated. Trump is actively destroying the dollar. Without the dollar the US isn't remotely as powerful as it is now.

@BlackwoodBrief The lunatic is de Villepin, if you have any understanding of geopolitics you get this. Unless you're a muslim ofc.

🔴 Former French Prime Minister Dominique de Villepin humiliates Trump and drags his dignity through the mud. 🔴 Stop this lunatic! Because of him, the price of a kilo of rice will reach 50 euros, meat 90 euros, and all food prices will increase. He wants to starve the world; he loves destruction. 🔴 The Strait of Hormuz was functioning 30 days ago, the world was stable, and people were happy until this lunatic came along and brought ruin to everyone, not just Iran. 🔴 Look at the oil prices now; they are insane, just like Trump. The whole world must unite and turn against him. France must join hands with Spain, Iran, Saudi Arabia, and everyone else to stand against Trump so we can salvage what can be salvaged. This man has been attacking Trump since the beginning of the events, and he is literally the most outspoken European who speaks the truth. 👏

@BishtSaheen I didn't know Americans were that politically illiterate. Well actually I knew but it's baffling. If the world stops buying and selling in dollars the US will be 10 times poorer. The US enjoyed prosperity because of its central positioning, if it can't stay the center it's over.

@TheGameVerse Except for the code.

Warframe director Megan Everett confirms no Gen-AI in their games: “Nothing in our games will ever be AI-generated.” Warframe and the upcoming Soulframe will remain fully human-crafted, prioritizing quality and authenticity in all game design. (via Gamespot)