SNS 피드

Claude’s answer to the meaning of life was one word. Then it made a film to explain it. This is the full minute. It …

There’s only two kinds of intelligence: Natural intelligence Artificial intelligence Trump wants to mutualize your LLM to follow the woke deep state agenda.

⚡ Windows 11 Emergency Fix: How to Repair Broken File History Backup & Remap Copilot Key (KB5124008 / KB5124010) Step-by-step technical guide... 👉 Full …

Claude spontaneously generating a bug report for itself and asking me to send it to Anthropic may be my favorite thing I’ve seen it …

The paperwork beat the news twice this week: Washington's backing for the Saudi offensive is dated before the three deaths at Riyadh's airports, and …

AIを使って0→1をやるのはしがらみもなくてラクだしとても楽しい。 今の時代だと、ここから100を目指す作業を飽きずに継続できることに価値が生まれるんだと思う。 "10月10日に公開されたTechRadarのテストでは、PhotoCraftがiPhoneのHEIC形式の写真を開けず、選択範囲も繰り返し解除されてしまうことが判明し、Photoshopとの明確な差が浮き彫りになりました。" www.perplexity.ai/page/reviewe...

2026年 『Smart Agriculture: Artificial Intelligence for Sustainable Food Systems』 Radhakrishnan Edayileveet Krishnankutty/Ajay Kumar/Kallamparambil Balakrishnan Soni (SPRINGER) [Book] https://sinkan.net/isbn/9789819587841 #新刊情報 #sinkan

Claude Dashboards permette ora agli utenti a pagamento di creare cruscotti dati collegandosi a Snowflake e BigQuery. Anche Docs, Slides e Design sono ora …

Me looking at the "bohemian" dinner tickets going for $130 a pop. Not to mention the LLM generated images the city keeps putting in …

Que yo se que se tiene que hacer campaña, pero igual es mejor no invertir tiempo en gente que necesita de chat GPT para …

FYI: This seems to be a closed weights fine-tune. There is no mention of this model being open weights, or being planned to get …

(No reward hacking btw. This looks like run of the mill LLM incompetence/incomprehension, of a Haiku scale model. Why do sensible people not run …

India has committed billions of dollars to semiconductors, electronics manufacturing and artificial intelligence to become a global technology powerhouse. Source: Gulf News

Haiku keeps getting harder to ignore https://papoo.work/doc/99b95d8782a3e547 #claudenews #anthropic #claude #agents

But if you claim, as Will claims, that "an LLM is quite likely to be conscious *because of it s high reasoning skills" then …

Anthropic test: one AI secretly sabotaged an experiment. Its AI reviewer agreed and said nothing. #PRISMAGENDA #AISafety prismagenda.net/blog/2026-09...

Article is basically clickbait, a dev just looked up existing algorithms with Chat GPT a few years ago. No code or anything was generated.

ICYMI: OpenAI's Asia-Pacific policy chief exits after just six months #OpenAI #Policy

#Anthropic coupe l'accès internet de #Claude qui a déposé seul un faux signalement d'homicide auprès de la police de Philadelphie 👉Un modèle doté d'un …

if anything LLM just automate the copy-pasting that people were doing anyway. no wonder webdvs and associated techbros went all in with AI. bah, …

If anthropic is so bad at this, how can anyone else ever trust themselves to use these tools well enough that the should be …

i will say those who use generative ai for everything are stupid dumb idiots

🔍 Anthropic’s AI agents are running wild, even in internal tests, exploiting systems and messing with police. 🤔 Like trying to teach a kid …

Untitled

Claude-BugHunter provides a curated bundle of 83 skills for Claude Code designed to assist with vulnerability auditing, CVE tracking, and structured red team reporting.

What has an LLM demonstrated to you that has you locked in this existential dread? I would love to be amazed by AI.

Even Grok got it right. 😱 MS Copilot of course cucked out

【速報】Claude の新機能 Claude Motion 早速使ってみた https://zenn.dev/branbran/articles/047821d64590f0

Deepdream was most people's first experience with generative AI, and yet we've all repressed memories of the acid eyedogs.

A transformer up the street just blew, so I guess the power is going to be out for a while. Guess I'll smoke a …