@nico16184

Yeah whoever builds the AI agent that curates the entire internet for you and presents you with content that is most likely to make your life better is going to make a lot of money. The algorithms are so strong, so potent, the only way to counteract it is an AI that acts in

Holy S*** - I tried Claude Cowork and it works like a charm. It works exactly as I imagined a computer agent should. It cleans up, sorts, deletes, renames, and so much more. And that's just the surface. The initial tests have left me deeply impressed. A dream come true.

My overall opinion on vibe coding has changed in the past year. I used to think there would be a a niche of vibe coders, similar to indie hackers. But I’m not sure anymore. I think within one year, vibe coding will just be a mandatory skill in business.

Anthropic's @_sholtodouglas says Dario Amodei's internal communication style is producing a compendium of essays in the company's Slack that will have effectively charted the history of AGI once we get there. "He has a really, really cool communication style. He quite frequently

I underestimated how emotional the impact of AI would be. For a decade, I was depressed about work. The best part of business is manifesting an idea. Seeing a problem, then fixing it for yourself and others. The worst part of business is trying to herd cats. Motivating the

Spend an hour reading this weekend and I think you’ll know more about robotics than 99% of people, including some people who invest in robotics.

startup idea for you: start an AI ads agency someone will make $50M+ in this space

ELON JUST PUT GROK INSIDE A TESLA CYBERTRUCK - AND THE OWNER SAYS IT SHOULD BE ILLEGAL A man casually talks to his Cybertruck like it’s a human - asks for food stops, coffee breaks, timing, and a fun NYC destination - and the truck plans the entire road trip instantly. No

Bilan de deux semaines intensives de Claude Code Max: - énorme FOMO en permanence, - j’ai testé des trucs ultra yolo («  connecte toi en root sur le serveur distant et désactive l’ipv6 ») - je suis persuadé que par moment il nous sert du Sonnet (et non du Opus) - parfois il est

The creator of Claude Code just showed you what an early AGI workflow actually looks like. Fifteen parallel instances, each functioning as an autonomous worker executing tasks unsupervised for hours. Starting sessions from your phone in the morning, checking output later in the

Tomorrow is day one of my "Vibe Coding for Business" series. As a founder in charge of growth i'm subscribed to over 20 different scattered software platforms. I'm convinced I can replace 90-95% of these apps with my own internal tools that I vibecode. Then, at the end of

until opus 4.6 or 5 drops, then codex 6, then gemini 4, then deepseek v4, and then the next wave after that the real point is how fast this is all moving at this pace, the next ".1" update is probably going to be a better coder than whatever you're using today

"What's the move tonight?" Brother, you're 32. Go do some squats and Claude code your way to hyperintelligence.

XAI PLANS TO FLOOD THE WORLD WITH MILLIONS OF DIGITAL HUMAN CLONES xAI is building "human emulators" that mimic keyboard clicks mouse moves and screen decisions to automate any digital job a human does right now. Just like Optimus handles physical tasks 24/7 at low cost, these

Cowork is now available for Pro

Claude Cowork gives you a peek into the future of knowledge work. Here's a demo of using Box as a cloud filesystem via MCP for Claude to leverage content as context to generate new slides locally. The future is having agents running in the background doing work for you.

Anthropic: Introducing Opus 4.5, the best model in the world for code & agents. We also made Claude Code more convenient for you. Btw, here’s Claude Cowork (it’s like Claude code for all your other knowledge work) OpenAI: Pay us or we’re putting ads in your chats

Made this plant persona (I call him Tiny) using @rive_app 's data binding, GPT and some hardware (ESP32 and a few sensors like touch, mositure, hum, light, temp). Most of the logic is on my client side rn. But will try and see how far I can take the logic part with scripting

Was true 16 months ago. Now it’s almost true… change 45 to 10

Using vibe coding tools for slide decks. Great idea! What are other uses like this? Going to spend my weekend making Claude Code Skills for each of these use cases.

Incredible things happening with Grok voice mode in Teslas I love when it starts mocking him at the end

claude code is having a moment it's looking like whoever wins vibe coding wins how you work on your computer @openai has some catching up to do:

Marc Andreessen: “I think Substack could be 1000 times the size of the existing content industry.” “Nobody ever asked for a Macintosh. Nobody ever asked for an iPhone. These things had to be designed and built, provided on the supply side before the demand materialized. And

Claude Cowork is now available to Pro subscribers.

When my students were creating initial demos with Claude Code & Antigravity, the AI would often spontaneously decide to do Wizard of Oz demos. The AI would build an interface, but not underlying logic. Code would (live!) run the interface behind the scenes to make it look working

Hired an e-girl to promote creatine

if you’re just discovering Claude Code for non-technical tasks, you’re in good company. in june told i told @lennysan it was the most underrated tool for non-coders:

god i love running an ai native business

They openly discuss their conspiracy to commit fraud and steal the charity

Check this out, this is gold (full resolution image below) Full end-to-end character creation workflow on @Scenario_gg Workflows - brainstorming to final output - all in one. Input a style, character description, and ref images. From there, everything runs automatically...

People getting tricked by a fake AI influencer. Welcome to 2026.

Demis Hassabis says that as we approach AGI, energy is effectively synonymous with intelligence AI will solve its own energy constraints by designing new materials, optimizing infrastructure and unlocking breakthroughs like nuclear fusion and room-temperature superconductors

Seeing 160 IQ polymath scientists who previously barely coded become demigods overnight with Claude code. The amount of innovation coming our way will be truly breathtaking. This is what the acceleration is supposed to feel like.

Elon Musk told an @xAI employee that he'd get a free Cybertruck if he could get a training run going within 24 hrs. "Elon's like "you can get a Cybertruck tonight if you can get a training run on these GPUs within 24 hours." We were training that night. He got the Cybertruck."

Next iteration of the Vibe Coder's Keyboard - now with Voice Mode!

we just hired our 25 founding engineers

"Claude Code, make me a version of the 1984 abandonware Apple IIe game Rescue Raiders, look it up" (A few pieces of feedback later) https://rescue-raiders-game.netlify.app

Anthropic Pavel Izmailov says AI won't make independent discoveries in the life sciences yet because it cannot run physical experiments But in mathematics, where models will likely master formal proofs and languages AI will soon become better than humans at proving technical

Anthropic keeps inventing new approaches to how agents work that then get adopted universally. First MCPs now Skills (which are an excellent tool). Also good to see fast adoption of good standards.

BREAKING: Mira Murati's Thinking Machines employees are all going back to OpenAI >3 co-founders rejoined OAI yesterday >more employees are joining today >50% of founding technical leadership now gone so over for thinking machines

The future is having 20 different livestreams with 20 different fake AI identities you've created to appeal to 20 different niches of people to sell 1 single product all at the same time. Buckle up!!

The work systems that were built to manage human teams using well-understood processes and procedures are going to fall apart in all sorts of ways when AI agents are brought into the loops without changes.

MALBORO DAKAR T5 → Concept-to-Photoshoot. With Weavy + Nano Banana Pro + Kling 01. (Full Workflow in thread) PROCESS: 01. Generate base T5 model. 02. Apply design to car render. 03. Iterate render shots. 04. Bring to life with NBP. 05. Apply drone aesthetic.

Claude Cowork is incredible Yesterday it saved me thousands of dollars and HOURS of my time I had it completely manage my finances while also planning my business and writing drafts of my content If you haven't used it yet, you need to. Here's your complete beginner's guide:

We're only 15 days into the new year. - In 15 days, 3 Erdos problems were solved (GPT-5.2 Pro), - Grok 4.20 (pre-access) found a new Bellman function for one of the problems Prof. Vanisvili was working on, - The CEO of Cursor used coordinated hundreds of GPT-5.2 agents to

BREAKING: Google Research just dropped the textbook killer. Its called "Learn Your Way" and it uses LearnLM to transform any PDF into 5 personalized learning formats. Students using it scored 78% vs 67% on retention tests. The education revolution is here.

Learn to Claude.

its so over..

Holy S***: CEO of Cursor said they coordinated hundreds of GPT-5.2 agents to autonomously build a browser from scratch in 1 week "We built a browser with GPT-5.2 in Cursor. It ran uninterrupted for one week. It's 3M+ lines of code across thousands of files. The rendering engine

Kling Motion Control is going crazy viral. Upload video + image → full-body motion transfer Kling Motion Control + all Kling models (2.6 / 2.5 / O1) UNLIMITED for 30 days on the Higgsfield yearly plan. Wild examples + how to try: #ad

Klarna CEO @klarnaseb never learned how to code. But after a summer using @cursor_ai, he now brings working prototypes to meetings instead of asking engineers to interpret his ideas: “Because I'm not an engineer, I've been vibe coding my whole life.” “I come into the meeting

Someone recorded a couple asleep in their @Tesla. The funny thing is, the sleeping couple is statistically safer than the driver filming on their phone while driving in the rain.

AI video is now good enough for longform content that looks completely real. This YouTube video has 3.6M views. The elderly person giving life advice doesn't exist. Not a single comment notices

Exactly how China wants it. Our naval resources are spread thin from Venezuela to Iran. Perfect opportunity for China to swoop in and take Taiwan

Berkshire Hathaway is sitting on $382B in cash. Before retiring at the end of 2025, Warren Buffett was hunting for a $100B+ deal but couldn’t find anything on the stock market: “I’m willing to spend $100B this afternoon. I’d rather have $100B and a really good business at a

We built a browser with GPT-5.2 in Cursor. It ran uninterrupted for one week. It's 3M+ lines of code across thousands of files. The rendering engine is from-scratch in Rust with HTML parsing, CSS cascade, layout, text shaping, paint, and a custom JS VM. It *kind of* works! It

I made tutorial on the "180 Degree Rule" for AI Filmmaking when creating dialogue scenes. Watch here: https://youtu.be/0VxTJYZ2Hcc?si=vLkSSC6Xm3_siYyb…

Gemini 3 commence à me saouler, j'ai l'impression qu'il est juste bon en data interne mais en raisonnement Opus 4.5 est largement au dessus, et je parle pour le quotidien hein pas que le code. Pour Opus il manquerai juste de l'ocr au niveau de gemini et encore plus de data

Teaching an experimental class for MBAs on “vibefounding,” the students have four days to come up and launch a company. More on this eventually, but quick observations: 1) I have taught entrepreneurship for over a decade. Everything they are doing in four days would have taken a

okay, so @vercel's new agent browser is pretty stinkin great. https://github.com/vercel-labs/agent-browser… substantially faster than playwright and is clearly optimized for AI use. well done, folks!

ELON MUSK: "I'M NOT REALLY A BUSINESSMAN, I'M AN ENGINEER" I'm sure there are probably lots of analysts on Wall Street who would agree that I'm not a businessman I'm an engineer. I like engineering and design I don't look at things from a rank-ordered list of priorities based

I have good news and bad news. 1. Good news for folks in Southeast Asia who will start making big bucks online. 2. Bad news for Western Instagram influencers and OnlyFans girls: you'll need a real job soon. Are you entertained?

this is so cool

Il y a quelques jours, Oxford Economics a publié un rapport sur le travail et l'IA dont le principal constat est que les entreprises ne remplacent pas les travailleurs par l’IA à grande échelle. Les chercheurs à la base de ce rapport soupçonnent que les entreprises "déguisent des

Seedance 1.5 Pro https://app.scenario.com/models/model_bytedance-seedance-1-5-pro…

Fun experiment: you can now automate content production with Claude Cowork. I gave Claude a browser and asked it to: 1) Find a NeurIPS best paper 2) Write a thread on it 3) Generate graphics on Krea 4) Validate its work with ChatGPT. Shockingly, it worked! How to do it

One of the more "boring" overlooked markets AI is completed upending is user research. Companies like Qualtrics ($12.5B), Medallia ($6.4B) and SurveyMonkey ($1.5B) have dominated for a long time. But now, we can infinitely scale and process interviews. Listen Labs, for example,

Anthropic: here’s a literal AGI that will be your best friend and help you make anything you dream of in a prompt. Also Anthropic: our iOS app is extremely buggy and it will fail and throw an error on every 3rd message Is this their plan to avoid fast takeoff?

Luc Ferry tacle Luc Julia et son histoire de ChatGPT qui ferait plein d'erreurs sur Victor Hugo : "Je lui ai fait faire 70 pages de biographie de Victor Hugo, je suis un fan absolu de Victor Hugo, y a pas une virgule à changer, pas une erreur, ni de date ni intellectuellement".

Un truc que j'ai mis bcp trop longtemps à capter avec Claude Code - automatiser la mise à jour de ton fichier claude . md - automatiser le test des dernières modif faites Via un fichier .sh "validator", qui joue un script avant chaque git push Intérêt : avoir une doc toujours

And YES - it's not perfect. But it's the first try. It's made in one click. And it's the worst it will ever be.

This was made in one. single. click. - No install - No dependencies - No broken nodes or library issues. - 100% flexible (image, video, audio... and more) We're so back.

Only one side had to build a wall to keep people from escaping. Many people risk their life on a tiny boat to get from Cuba to America, but none go the other way.

Marc Andreessen on AI: “This is clearly bigger than the Internet” “This is the biggest technological revolution of my life,” a16z co-founder Marc Andreessen begins. “This is clearly bigger than the Internet. The comps on this are things like the microprocessor, the steam engine,

“My bet is that over the next 18 months, OpenAI runs out of money. OpenAI failure wouldn’t be an indictment of A.I. It would be merely the end of the most hype-driven builder of it.”

KYC Video Verification is officially dead

Claude Code probably saved like 1 billion hours worth of engineering productivity already

BREAKING : Meta officially launches new initiative at 'Meta Superintelligence Labs' (MSL) under Chief AI Officer Alexandr Wang. In his first act of "efficiency," Wang has reportedly replaced the entire executive suite with 100% optimized versions of himself with Claude Cowork

Cool

omg.. Stanford AI Lab just dropped WonderZoom you can generate 3D world from a single image and.. infinitely zoom into any tiny details..

Had Claude Code build a little plugin that visualizes the work Claude Code is doing as agents working in an office, with agents doing work and passing information to each other. New subagents are hired, they acquire skills, and they turn in completed work. Fun start.

Money makes a big difference: Trainers, chefs, 300 days off a year…. And a good plastic surgeon.

You will either be delegating to Claude. Or someone else's Claude will be delegating to you. Be the first one.

Performance driven AI videos will dominate 2026 This workflow gives you maximum control: 1. Use this 3x3 dialogue prompt in Nano Banana Pro to create different shots of your scene 2. Animate via Kling AI 2.6 Motion Control to transfer your own acting & lip sync Prompt below:

I recruited the best vibe coders I know for an all-day livestream on January 22. @OfficialLoganK, @ryancarson, @bentossell, @fkpxls, @nateliason, @ashebytes, and the @every team—building live, sharing exactly how they work. Free and open to everyone: https://every.to/go-agent-native

I forked Claude's pptx skill to make Claude Code generate gorgeous presentations Claude can generate presentations using the pptx skill from @AnthropicAI but its output can be quite bland. Especially if you're used to the high aesthetic standards of @every. So I played around

Claude Cowork was a collective awakening. Despite it being 10x worse than @ManusAI, Claude somehow captured the awakening. Over the next 6months we are going to see individuals create agentic apps that we never thought were possible. Some of these people will be fully

ICYMI: Yesterday @AnthropicAI dropped Claude Cowork—essentially Claude Code for everyone, not just engineers—and we livestreamed a Vibe Check at @every as soon as it dropped. Felix Rieseberg (@felixrieseberg), a product engineer at Anthropic, joined us midstream as a surprise

This is the first example I've seen where subscribing to an AI agent gets you access to proprietary, paid data. SimilarWeb offers just 1 month of data on a free plan. On a Manus plan (free or paid), you can now get 12 months of SimilarWeb data.

I had AI break down global politics like it was Game of Thrones... here’s how that went! and I'll walk you through how I did it on Freepik step by step below, prompts included

WTF is a Claude Skill anyway? Here's my simplest explanation: It's just a folder with text files that AI can trigger based on a given task from any chat: 1. Skill(dot)md A required file that tells AI when to trigger the skill and what to do. 2. Reference files Templates and

Etienne Klein en 2025 continue de rapporter ce "test au labo" (ça fait sérieux hein) en évitant de préciser (et en ignorant peut-être) qu'il n'a jamais fonctionné que sur la première version de ChatGPT publiée en 2022. En quoi est-ce encore instructif sur les LLM aujourd'hui ?

I have no special knowledge of the Google-Apple AI deal, but if the phone becomes a vessel for working with AI assistants, you probably want to own the assistant, not the phone. Also a sign that, until things hit a wall, even well-resourced firms can't catch up with the frontier

Anthropic just dropped Cowork, a Claude Code for the non-technical tasks It lets Claude work directly inside a folder on your computer: - Read, edit, create files - Create a spreadsheet from a pile of screenshots - Re-organize your downloads - Plan and execute tasks - With

still can’t believe I can do this with no coding bg at all, just chat with AI agent to add anything full tutorial and prompts below

If you've never coded in your life you can download Claude Code right now and by the end of the night have a full app 99% of people have have no idea You can't imagine the gold rush that's about to happen when people finally wake up

Absolutely mind boggling to stop and realise: 1) I barely code anymore, just chat with an AI. 2) It’s not slop. I’m still engineering. I don’t feel threatened. Feels like pair programming. 3) I’m enjoying it more than coding by hand. Truly wild time to be living through.

6 Hour Claude Code live stream in 12 days. For 3 hours I'll be talking about Claude Cowork, and the next era of agentic software. Will be joined by multiple guests who are Claude Code power users. Will be very fun. Link below.

claude wrote 100% of claude code claude code wrote 100% of claude cowork wild times we live in

cool at least ill still have two jobs

Just hired 8 founding engineers

When OpenAI launched this 3 months ago, people we're saying that "1000 startups died". This has zero users.

Claude's gaps in image creation and voice became less important when Claude Code just uses APIs to go to nano banana or ChatGPT imagegen whenever it needs a picture made, or ElevenLabs API when it needs a voice.

Claude Code with Opus-4.5 is basically AGI. Here's how to feel the AGI (with expert @carlvellotti): 11:04 - Setting Up MCPs 28:00 - Using Skills 38:00 - APIs 44:12 - Writing Docs 1:08:08 - GitHub Integration

Hi, I'm Felix and I work on Claude Cowork, bringing Claude Code closer to all kinds of knowledge work. It's an early and rough preview, please tag me in any feedback - we want to iterate very quickly and make it a little better every day.

A week and a half. That's how long it took @felixrieseberg and his team at @AnthropicAI to ship Claude Cowork.

the only jobs left after the singularity: podcaster degenerate gambler vibecoding influencer black market chinese peptide dealer etsy witch

Claude Cowork isn't chat—it's async AI for the rest of us. Ask your computer to do something. Walk away. Come back to results. Claude Code users know this feeling. Now non-technical folks get it too.

Send this to every normal person you know because normal people don’t want to touch a terminal and this will make them 100x more powerful and productive at whatever is they do

We’re live with @felixrieseberg from @AnthropicAI to talk Claude Cowork—Claude Code for non-technical users. Join us https://x.com/i/broadcasts/1nAJEELnLweJL…

New leaks to OpenAIs upcoming device: - It’s a new audio wearable, meant to replace AirPods. (aligns with The Informations leaks) - Codename: Sweetpea. - Look: metal “eggstone” + two little capsules behind the ear. - Tech: aiming for 2nm chip + maybe a custom chip for

Is Claude code really better than Claude in other agent harnesses? Which harness is best?

Grosse mise à jour sur https://gemini-design-mcp.com ! - J'ai refait tout le system prompt pour push Gemini a son max niveau créativité et design Avant / Après

Mark Cuban’s advice to people trying to get into AI is simple. help businesses implement. SMBs have budget but zero expertise. they’re not waiting for AGI. they’re waiting for someone to show them what’s possible. what’s stopping you from being the one they call?

What if you can go from a real-time conversation to a finished slide deck, website, or social media collateral, all in ONE flow? Introducing Meeting Minutes, a new feature that transforms your in-person discussions into structured notes you can act on immediately. Available

Wow, such a great open-source drop! Put Claude on steroids with 120+ scientific skills spanning maths, biology, chemistry, medicine, engineering, & more If you’ve ever wanted Claude to act like a research assistant, this gets very close. Free and open source. Repo in ↓

Complete Beginners Guide to Claude Code. > Intro to Claude Code > Easiest way to use Claude Code > Planning and Thinking > Good vs Bad Prompts > When things go wrong > Getting Better over time

There's almost 3k stars on the Ralph repo now. Getting some interesting PRs too - thanks everyone! If you want to get Ralph setup, just point your agent at this repo:

All you need to get rich in 2026 is 10,000 followers on X and a Claude Code subscription

I've spent the last 48 hours in Claude Code - as a non-technical person it's basically unlocked three very big things for me 1) The ability to interact with APIs generally - again, as a non-technical person one of the big barriers to running the business has been touching APIs.

Alibaba Qwen lead researcher, Justin Lin: "chinese companies are heavily limited by inference compute" hard to disagree with. the US leads in compute and keeps pulling ahead because it attracts more investment, which turns into more data centers and more compute but over the

Vous etes des monstres les gars J'ai ship Gemini Design MCP y'a 4 jours et les abonnements s'enchainent non stop, j'ai jamais vu ca A la base je l'ai build pour mon propre setup parce que j'en avais marre de switch de tool pour le front, et la ca devient un vrai produit mdrrrr

This open-source Neural Network Visualizer is an absolute game-changer, rendering activations in mesmerizing 3D

This is not sped up. So folks. conclusion still stands. Robots are coming for all jobs, even the ”safe” ones. It’s just a matter of time. it’s when, not if.

My annual MRI scan gives me a USB stick with the data, but you need this commercial windows software to open it. Ran Claude on the stick and asked it to make me a html based viewer tool. This looks... way better.

You can now use Claude as an orchestration agent for long-form video generation. From one text prompt, it can direct various tools and combine outputs - like Nano Banana Pro for images, Hailuo 2 for video, and ffmpeg to stitch. Watch it construct my bookstore x coffee shop

THE FASTEST WAY TO CHANGE YOUR LIFE IS TO MAKE SOMETHING PUBLIC THE INTERNET REWARDS PEOPLE WHO TRY OUT LOUD.

Mark Zuckerberg literally drops a 90-minute masterclass on AI, China, and AGI

"If I could summarize what I've learned about Claude Code, it's text files all the way down." Here's my new episode with @talraviv and @amankhan where we demoed 5 real AI workflows that we actually use, including: Writing strategy docs with Projects Prototyping features

Elon Musk estimates building a Mars city would cost over $1,000 trillion. But with a 1000X rocket tech improvement, costs could drop to $1 trillion over 40 years, or under $25 billion per year

This tweet needs millions of views. Seriously. You have no idea how big and important this is.

3 math problems - unsolved for decades - solved in the last few days

This is wild. GPT-5.2 Pro just cracked its third Erdős problem in days. Not benchmarks but real open math, accepted by Terence Tao. OpenAI absolutely cooked. This is the first genuinely PhD-grade AI we’ve seen, doing original science humans hadn’t solved. We crossed a

And Erdos problem number 3 has been solved by GPT-5.2 pro (#379). It was accepted by Terence Tao. I believe the last few days have been the "lift-off" moment for AI science.

"Robots will never replace skilled trades." The robot…

http://Grokipedia.com will soon far exceed Wokipedia

opus 4.5 is better for a polished frontend UI for almost everything else, codex is much stronger opus also seems to hit a wall on harder tasks and sometimes rolls back what it did codex usually keeps going opus is faster, but codex ends up producing higher-quality, working code

This is awesome, I love it. Gesture interactive particles using Opus 4.5, Three.js, and a media pipeline.

a saas company wants to charge me $900/year to renew. let's just clone their business instead.

The Ralph Wiggum Claude Code plugin is taking over X and for good reason It allows Claude to work for DAYS by itself I've been able to one shot multiple apps and features with it In this video I cover how Ralph Wiggum works, how to install it, and how to build with it:

Stephen Wolfram says AI won't end work, but it will make learning low-level mechanics obsolete We moved from assembler to high-level languages, removing the need for humans to understand the machine's plumbing "you don't have to go to the lower levels... it's no longer useful

Kling’s text-to-video capabilities still blow my mind every time.

every failed VC-backed consumer app from 2015-2022 is now a perfect $1-10M solo business just waiting for someone with claude code etc to rebuild it in a weekend and unleash a flood of UGC creators behind it go find those "we're sunsetting" emails like there is no tomorrow

This looks impressive. Anime is predestined to be the first genre to be largely AI-generated. We will likely see the first full-length, completely AI-generated anime by the end of 2026.

Claude for Chrome is absolutely insane with Opus 4.5 IMO it’s better than a browser - it’s the best agent I’ve tried so far

GPT-5.2 Pro continues to do the most impressive things on hard problems, but it does so with almost no visibility into what it is actually doing. The thinking trace is often unrelated to the final result, the tool use is unclear. No explainability, just remarkably good answers.

the future of AI orchestration is 3D

Want to know how to vibe code games like these? - ThreeJS Claude Code / Codex Agent Skills - Nano Bana Pro - Game Design + Technical Design Specs - Opus 4.5 / Gpt 5.2 Here's my FULL step-by-step workflow Source code + agent .mds + playable links in reply

prompt 手势交互式粒子，最重要的是你要强调你要用到three.js。

Anthropic is so powerful that everyone uses Claude. Now they're using their dominance to block xAI from using Claude via the cursor. That was 4D Chess play by Dario. If you have the best coding model, you dominate everything.

BREAKING: Google Gemini launched a new feature called Guided Learning. You can now use it to learn literally anything, step by step, like a personal tutor. Here’s how to access it

Building space stations for humans is going to be such an exciting step in the development of humanity. Grok Imagine, of course

INTRODUCING: @doanythingapp Do Anything Agents are a totally new kind of agent They: - work independently for weeks/months+ - have their own email - self manage entire projects - can use almost any tool on the web Today the alpha opens to the public. Here's how they work:

Jack Clark @jackclarkSF co-founder of Anthropic @claudeai, Dwarkesh Patel @dwarkeshpodcast, and I had a good discussion, and it has been published, free to all, at the link below. I greatly admire both Jack and Dwarkesh and was happy to participate. The AI revolution is here.

For what its worth, Claude Code does seem like a genuine breakthrough moment for many folks whose work I respect (& to me as well). Great AI + great agentic harness resulted in something better than either alone & feels like a threshold crossed. Other labs are close to this, too.

if you want to learn about agent-native apps come to my livestream with @awilkinson in an hour 1:30 PM EST here on X

NEW: i wrote a complete technical guide to building agent-native software (co-authored with claude) it covers: - the five pillars of agent native design (parity, granularity, composability, emergent capability, self-improvement) - files as the universal interface - agent

It's time to build.

Approximately 1/3 of my X feed is people gushing about Claude code. I’m already an intensive ChatGPT user so I am open minded. And I will try it. I can't help but wonder: 1. Why do most of these posts sound like they were written by their AI? 2. Is this a viral marketing

Huge: The next generation of the DeepSeek series is scheduled for release in early February and is expected to outperform competitors like Anthropic's Claude in coding! Is this really the moment when China overtakes the USA?

This billionaire literally explains where OpenAI, Apple, and Anthropic are actually headed in the AI race

我用 Opus 4.5 编写了一个手势交互式粒子。 你可以用双手流畅的控制粒子 (three.js + mediapipe) ，大家可以试一下

Terence Tao says the current business model of big AI companies relies on hype to justify their massive valuations The situation is similar to the dot-com boom, where the tech was real, but the early promises were inflated "hopefully we'll reach a more realistic equilibrium in

I built a Batman game in ONE EVENING using only AI prompts No game engine. No coding skills. Just @AnthropicAI Claude Code + @Google Nano Banana for sprites. Crazy part? My first game on Amstrad CPC was Batman. Full circle. I. Don't. Know. How. To. Code. As @arohan said:

Claude Code for Chrome is really something else. I haven't used Google Analytics in a minute, not even sure what I needed and the product has changed so drastically over the last few years. One prompt and I got some nice dashboards to get me going.

I've increased your tokens for all free plans :)

this is the CLEAREST explanation of "ralph wiggum" on the internet and how you can use it with claude code/amp/etc and have AI agents build software for you 24/7 even while you sleep

our time has come

My 2nd indispensable tool in my setup Claude Code is Gemini Design MCP The principle is simple, you delegate ALL the frontend and design to Gemini while your main agent handles the logic, the database, the backend. Whether it's Claude Code, Codex or Cursor it works - You

One X feature that would be extremely useful in the age of AI (and would actually make the X articles feature make more sense) is to let me download my bookmarked posts The fact that there is no way to do so officially is a huge gap, especially as older bookmarks can't be viewed

some 4k goodness nano banana pro + kling 2.6 + topaz video upscaler all on @freepik

The setup: VPS used solely for Claude launched in a tmux session to keep it running 24/7. It has sudo access and can do ANYTHING on the server. The server, my Mac, and my phone are on the same Tailscale network. I connect via the Termius app on the phone and SSH connection to

I only use this now. No other chat anymore.

ELON: MOVE SKILLS BETWEEN WORLDS AND YOU BECOME SUPERMAN Elon says the real advantage comes from solving problems across completely different arenas. When knowledge crosses fields, difficulty collapses. What feels ordinary in one industry becomes overwhelming leverage in

this was impossible before AI made live action Thundercats in a few hours, under $100

Introducing CallMe, a minimal plugin that lets Claude Code call you on the phone. Start a task, walk away. Your phone/watch rings when Claude is done, stuck, or needs a decision. Free & open source (MIT). Underlying API costs are cents per minute of call.

Marc Andreessen: "I'm reading every spare minute that I have" Expanding on his information diet, the a16z cofounder says AirPods have been the single biggest technological leap: "They're the unlock for me for audiobooks, podcasts, and interviews. I'm doing audio content

$10M is easy if you leave this bubble If you know what Claude Code, clipping, or MCPs are- you're in a bubble X is a magical corner of the internet. Take what you learn here and apply it to industries that aren't paying attention.

this is weird, my claude code won't start

the timeline is full of people using Claude Code to generate ads, recover corrupted video, make landing pages, manage business texts… the narrow view of coding agents is that they primarily decrease the cost of software development the ambitious view is that almost any

2026 feels like a generational lock in moment daily exposure to AI tools is probably the single greatest thing you can do now: launch 7 apps in a year, double down on 1, become a weapon at your job, get promoted, find freedom, life changed right now matters so much

My 2026 AI predictions podcast with @reidhoffman—LinkedIn cofounder, Microsoft board member, and former OpenAI board member: Reid’s spiciest predictions: - If you’re not recording every single meeting and using agents to amplify your work process, it’s going to feel like using a

We've been pushing the boundaries of what an AI agent can do. Here’s a thread of real use cases from our internal demos that showcase things only Manus can do. 5 side-by-side comparison videos to show you what a real AI agent can do.

we have entered the singularity. it’s here. hold on tight because this year is going to be wild.

Motion control just took a quantum leap. Kling 2.6 Motion Control doesn’t just generate video, it mirrors human nuance with precise body mechanics and true 1:1 action copying. The benchmarks aren’t even close. See it in action

Claude Code is just wild. This entire Hermès-inspired ad was generated autonomously: → script, voice, visuals, music, branding... everything. Claude Opus 4.5 wrote the script, orchestrated ElevenLabs, Google Veo 3, ffmpeg, downloaded music, and handled production end to end

Everyone who tried Claude Code over the holidays and has been thinking deeply about this for 7 days: All white collar jobs are toast. It is over

- The world's best devs are blown away by the evolutions of Opus 4.5 and have integrated it into their workflow (at worst in red team for an extra layer of review) - Claude Code is basically coded by Claude Code - Google's engineers recently admitted that Claude Code had

Developer uses Claude Code and has an existential crisis We will probably see much more posts like this.

This guy literally shows how to use Claude code end-to-end in 20 mins

Elon says Grok 5 will have a perfect score or a “near perfect” score on Humanity’s Last Exam

Lmao. These Maduro AI videos are getting out of hand.

Claude Code is one of the worst named products. It's not just for code. Watch this Hermès ad. 30 seconds. 8 shots. Voice. Music. Even text branding burned in with ffmpeg. Clopus 4.5 wrote a script, orchestrated ElevenLabs, Google Veo 3, downloaded music and made this from

Amazing post on vibe coding. “the thing i got wrong at first is thinking vibe coding is about learning to code and... it's not! it's about learning to communicate the skill isn't python or javascript or whatever, the skill is clarity … the problem wasn't Claude. the problem

Elon Musk says the AI community underestimates "intelligence density" potential by two orders of magnitude We could extract 100x more intelligence per gigabyte, per watt, per transistor — just from algorithmic improvements alone "we're looking at 10x smarter AI every year"

This interaction between sama and elon is the most wholesome thing ive ever seen.

Opus 4.5 is really much more helpful at writing than GPT-5.2 and Gemini 3. It seems under-discussed how big the difference is.

if you want to be an entrepreneur today, i’d take $200k in claude credits over 4 years of random college credits all day

Elon Predicts Universal High Income and Social Unrest. And It Terrifies Me. Elon Musk wasn’t being edgy here. He was being honest. And honestly, it scared me to no end because I think he’s right. The transition to a post-labor, post-scarcity society will likely be the hardest

If want to use Claude Code but are don't like the terminal interface, you can now use local Claude Code from Claude Desktop! To do so: 1. download Claude Desktop 2. open the sidebar and click 'Code' toggle 3. select the folder that you want Claude Code to have access to 4.

create multiple cinematic shots with a streamlined workflow : from Midjourney 2D concept art to full-length movies with Nano Banana nodes on Freepik Spaces

Holy Sh1t, LeCun has a total colapse

Elon Musk: Optimus will be the world's best surgeon within three years.

I think if you use Claude Code and Gemini Nano Banana Pro (as mcp) you can basically make all the video games you want Nano Banana Pro can make sprites Claude code can write chroma keying Claude can generate music / story lines et. Concept from Nano Banana.

we're in the weird middle phase of AI right now: when chatgpt launched, it felt like a toy at first, but after gpt-4 it started helping with real tasks for me, the real breakthrough was reasoning/test-time compute, when openai launched its first reasoning model, o1, in 2024

DeepMind's mission: "Step one, solve intelligence; step two, use it to solve everything else."

Vibe coded a game character selection screen Everything here was made with AI tools Nano Banana: character design + UI Tencent Hunyuan3D: image to 3D Gemini Pro: UI More details ↓

This AI is wild. Higgsfield Relight lets you change the light after the shot 8 examples + how to try: 1. Change light scheme #ad

At the start of this year, your timeline is going to be filled with posts about Claude code: that's normal. It's a major paradigm shift, and I'm weighing my words. It's a leap as powerful as the arrival of ChatGPT in 2022. The business utility of AI is the product of 4 factors:

It's more like that

I've started reading Mr. Phi's book on the subject. The evolution over the last 3 years is dizzying for LLMs. But what is going to happen with the AIs that evolve in or interact with our environment? Probably something unexpected and bewildering!

I built an app with Claude Opus 4.5, then animated it with this new tool I found... This new iOS design workflow let me animate a character for my onboarding. Here it is: 1. build the app using Claude Code within @vibecodeapp 2. entered image of my mascot into

Emad Mostaque says if your job is on a screen, AI will do it better for pennies within two years 2026 will be when economic agents are benchmarked in dollars By 2027-28, companies will deploy "digital doubles" that run nonstop and don't repeat mistakes "it's a new economic

hard to believe that opus 4.5 was released 10 days ago.

i thought the view was surreal, but i looked twice and found out that it was just satya Nadella and marc andreessen

I just showed the team our 2026 roadmap. I got so psyched that I wanted to share it with you too. Here's our full 2026 strategy.

This is how I code now about half the time From my phone with Termius coding on my VPS and writing requests into Claude Code like a goon

"Claude Code just SAVED my marriage!" I gave Claude Code 512 GB of HEAVILY corrupted wedding footage from a close friend of mine, and within 10 minutes, recovered it all. Two days ago, I got some bad news from a buddy telling me how the only copy of their wedding videos was

His statement is already wrong and it’s less than a week into 2026

Dive into the fun of Kling 2.6 Motion Control—the trend we're defining. Share your creations and tag us!

Bonne année 2026 à tous :) Une année qui s'annonce époustouflante sur le plan de l'IA ! Merci @nico16184 pour la pépite Attention aux pièges Sur le plan du body language, un basculement me frappe : nous sommes passés de visages HYPER-statiques (raideur, figement) à des

Wallstreet is so cooked.. this new AI agent can analyse real-time stock market data 24/7 autonomously and tell you the best time to buy or sell.. it's like having a personal finance expert in pocket here's how it works:

Geoffrey Hinton says mathematics is a closed system, so AIs can play it like a game. They can pose problems to themselves, test proofs, and learn from what works, without relying on human examples. “I think AI will get much better at mathematics than people, maybe in the next

Yes, Claude 4.5 Opus and GPT-5.2 pro were the tipping point, the moment when everyone realized that something had changed forever. From here on out, things will only get faster.

Alexei Grinbaum (CEA) explains that LLMs are now a genuine medical aid to individuals. The oncologist Jean-Emmanuel Bibault recounts the recent story of an American patient who was turned away from the emergency room, then finally accepted and correctly diagnosed thanks to the

“AI is slop” they say

Anthropic's new AI model can digest financial statements, analyze them, and put together an investment memo within matter of seconds Junior roles in finance will soon become a commodity. AI is coming for their job faster than they think

Chamath: Two terms you need to pay attention to in AI are Prefill and Decode “There's two terms that I think you're going to hear a ton about over these next few years.” “The first term is prefill, and the next is decode.” “What prefill and decode are, are two very distinct

Hot take: We spent 2025 teaching AI to execute. I feel like "can AI do this reasonably well?" is mostly solved and will continue to get much better. What I'm interested in for 2026 is AI that understands timing and context and acts proactively without being prompted.

Had a lovely dinner last night with @POTUS and @FLOTUS. 2026 is going to be amazing!

I think you missed the main ideas. - The basic premise of JEPA is that training by reconstructio/prediction in input space is evil (or counterproductive). The details are almost always unpredictable. Hence prediction must take place in representation space, where unpredictable

Claude Code with Opus 4.5 is a watershed moment, moving software creation from an artisanal, craftsman activity to a true industrial process. It’s the Gutenberg press. The sewing machine. The photo camera.

I’ve taken a little break over the holidays from uploading (new video next week). Instead, I’ve built 5 different apps to solve little bottlenecks in my business. I’m addicted to building. I even built an iPhone app with @vibecodeapp that I use daily and am considering selling

This is definitely how I've felt these past two weeks. Getting AGI shock every day. Chat with the Lindy team on Friday: "AI coding is solved at this point." If you think you know what's gonna happen next, you're very confused.

100%. I spent the break writing an essay about how the fundamental variables that sustained human existence for 200,000 years have now shifted.

The Claude Code effect strikes again. Midjourney founder David: „ive done more personal coding projects over christmas break than i have in the last 10 years. its crazy“ And „Nothing is going to be the same anymore“ Anthropic achieved what OpenAI tried to achieve.

Fiona Fung currently supports the Claude Code team at Anthropic and was previously a Senior Director at Meta. She grew quickly through the ranks at Microsoft and Meta before joining Anthropic. I interviewed her about what she learned along the way. We discussed: • When to trade

I built an AI Factory that uses Claude Code agents to build apps while I sleep. This is an automated "assembly line" pipeline, all run by Claude Opus 4.5. Projects move through a Kanban-like system from Idea to Research, Architecture, Coding and Testing. Factory workers start

claude code will probably make 50,000 people millionaires if not more

Anthropic just killed thousands of AI apps Claude can now create and edits Excel, PowerPoint, and PDF files directly in your browser.

Jack Ryan clip about Venezuela gets viral amids capture of Nicolàs Maduro.

We have entered the Singularity

The user is totally empowered now, we need more endpoints to direct our tokens I have been doing game dev vibe coding over the holidays and its next level with Nano Banana and Opus 4.5 Integrating these systems into the Cursor for Unity/Unreal is fast approaching

ive done more personal coding projects over christmas break than i have in the last 10 years. its crazy. i can sense the limitations, but i *know* nothing is going to be the same anymore.

They've already pulled the anime about Maduro's detention, lol.

Andrew Huberman's powerful message: Your brain can change at any age — but it takes deliberate effort "Neuroplasticity is real. As a child, passive experience shapes the brain. After ~25, change requires a major shift in neurochemical state — alertness + focus. High-intensity

ive been hanging out with founders under 22 lately living in sf/nyc and they're built different. my observations of these young founders getting rich with AI: 1. these kids grew up watching YT creators flexing Porsches and private jets from their bedrooms. but when they looked

Wow... Insane Claude Code hack While using Claude Code on your phone, you can just tell it to create a Skill... And Claude Code will use it when the time is right... Prompt / Instructions below

Claude Code isn’t just for coding. I fed it my raw DNA data from an ancestry test and used it to find health related genes I should keep an eye on. The file is massive, but its ability to search what matters makes it possible.

WHY VENEZUELAN OIL MATTERS: THE U.S. JUST FLIPPED THE ENERGY CHESSBOARD The U.S. has effectively gained access to nearly 1 million barrels per day of Venezuelan heavy crude, oil that was previously flowing to China. That crude is uniquely suited for refineries in Texas and

I bet we have a unified foundational theory of physics in the next year or two Will be very cool Will be similar to thermodynamics or special relativity, a top down theory of principle, not a constructive theory done by fitting loads of lagrangians

And off we go to our umpteenth CES in Las Vegas with the pals @Fsorel and @simottel

Learning Claude is the best upskill this year. Nothing comes close.

It's over.

My predictions for 2026 1 Europe will become aware of its immense technological lag and its colonization but will be unable to act: in Brussels there are lawyers and bureaucrats but no entrepreneurs! 2 The National Education system and universities will be panicked by AI: they

Google engineer: "I'm not joking and this isn't funny." "Claude Code generated what we built last year in an hour." REMINDER: "Self-improving Al is 'really scary' ... Things could get out of control very quickly." -Sam Altman (2015)

“There’s nothing artificial about AI. It’s inspired by people. It’s created by people. Everybody should care about AI because it's going to impact your individual life, your society and the future generation.” ~ Dr Fei-Fei Li (@drfeifei )

here's my explanation for why people use Claude Code instead of Cursor

When you start using 3-4 agents and more simultaneously, it becomes complex to monitor in real time all the changes and to guide them correctly. It's like having 3-4 full-time employees; if you don't have visibility into their work, it's difficult to assign them the right tasks.

opus 4.5 is actually batshit insane this thing can do anything I ask it with some high-level system design guidance and a clear path for verifying itself and I've been coding for over a decade you'll soon see $1B+ one person companies 2026 is going to be for the books

Can't stop iterating on this ! Opus 4.5 + Claude Code to plan design polish GPT 5.2 + Codex CLI to implement the plan Result: - More weapons - Particle effects + visual VFX - Dynamic camera - Destructible items And here's a cool kill-cam at the end of the round

I'm not joking and this isn't funny. We have been trying to build distributed agent orchestrators at Google since last year. There are various options, not everyone is aligned... I gave Claude Code a description of the problem, it generated what we built last year in an hour.

You can do this with slash commands. Enabling it for skills also in the next release. Thanks for the suggestion!

I believe Elon & SpaceX will attempt to send Optimus to Mars aboard Starship in the Nov-Dec 2026 window. Elon told me there’s a 50/50 shot and if so, I will be there to witness it first hand with my family. It’s one of my goals for 2026. Go team humanity!

Anyone who thinks AI is a bubble has never used Claude Opus 4.5

Influencers are screwed. Nano banana pro is indistinguishable from reality

I hope this was helpful! What are your tips for using Claude Code? What do you want to hear about next?

I'm Boris and I created Claude Code. Lots of people have asked how I use Claude Code, so I wanted to show off my setup a bit. My setup might be surprisingly vanilla! Claude Code works great out of the box, so I personally don't customize it much. There is no one correct way to

Scam caller didn't stand a chance against her [ theafryknas]

"Many people have left, and many others will leave": Yann Le Cun, one of the fathers of AI, reflects on his departure from Meta and shares more about his future start-up https://l.bfmtv.com/0cB9

That's an interesting idea. How do you feel about this, folks?

Terence Tao, Fields Medalist mathematician, explains that neural networks do not understand concepts, but they uncover hidden patterns. In knot theory, a classical neural network discovered correlations between mathematical invariants no one had anticipated. It started as a

Just because something has never happened, doesn’t mean it can never happen. Humanity never flew, and then it did.

Elon Musk: My philosophy is that if you play, you play to win.

we've come a long way in just a year looking at the 2024 → 2025 "IQ test," only 1-2 top models were even brushing the 120+ range now, most leading models typically score between 125 and 130 IQ i know this test isn't a perfect way to judge LLMs, but even with that caveat, the

2030s will be the 2000s that we were promised in the 1970s.

If you want to master AI in 2026, this might be the best video to start with. Karpathy released a free 3.5-hour deep dive on how LLMs like ChatGPT actually work.

Mathematician Terence Tao says: "AI tools are now starting to solve the easiest 'open' Erdős problems, and many turn out to have already been solved quietly in the literature the category of nominally open problems... most likely to have been solved in the literature" i'm not

What a year 2025! THANK YOU! On the content creation side, it's 3 million video views (and nearly 10M cumulatively across videos on all networks) (if we add text, we're nearly at 55M views anyway wow...) on projects that have always served me to question. - I coded AIs that

This billionaire literally breaks down how AI will reshape software, economics, and biology in under 1 hour

Instead of watching a 2 hour movie, watch this masterclass on the future of work

In retrospect, the articles mocking Dario’s prediction that 90% of code would be written by AI by September seem to be very misguided. He seems to have been only off by a couple months (if that).

we haven't fully realized what 2026 will look like these models/agents are already running autonomously for more than 10 hrs opus 4.5 can get through about 5 hours of human-estimated work, and gpt-5.1 code max is close to 3 hours with a 50% success rate task length is doubled

Before you vibe code, be interviewed by AI Here's how: - Download Claude Code (different from Claude Desktop) - Install it with your Command Line (terminal, powershell, wsl, etc) - Write a basic project spec file such as `Accounting software for YouTube creators` in

New episode w @AdamMarblestone on what the brain's secret sauce is: how do we learn so much from so little? Also, the answer to Ilya’s question: how does the genome encode desires for high level concepts that are only seen during lifetime? Turns out, they’re deeply connected

You are insufficiently astonished by what Claude can do with the right scaffolding. Here I asked it to generate an influencer video explaining LDL cholesterol and statins on a white board.

Congrats to @ManusAI team on their way to @metaai I wonder if @Microsoft or similar will snap up @genspark_ai… If you want to try the best open source version fully private it’s II-Agent by @ii_posts Lots more features to come, agents now give a big boost over raw models

Pluto is a planet.

Over 3 billion miles away, we arrive at Pluto (I am a planet, deal with it).

Great explanation from a Reddit user on why Claude Code is so appealing for non-coding knowledge work.

Probably not a coincidence.

Success isn’t sorted by intelligence. Dumb people with nerve end up successful. Smart people with nerve end up immensely successful. Smart people without nerve end up depressed, bitter, and loud. Balls matter more than brains.

Saying you are positive or negative “about AI” feels like a category problem. You can be very positive about AI & very negative about AI around exactly the same topic For example, early evidence on AI tutoring is incredibly promising, but its initial impact on education was bad.

.

There are broadly two ways people think about AGI and labour: Position A is where humans get fully substituted, which is usually advanced by parts of the AI commentariat. The argument is that if AGI is a scalable input that can do what workers do at lower cost, then the market

I now believe this also works the other way around - LLMs love autistic people, too. They often produce unexpected sequences of tokens. That must be somewhat thrilling to an LLM!

In 2026 the only way to accurate prediction how successful you’ll be is my measuring how much you use Claude Code. That’s it. That’s the only thing that matters.

Hot take, videcoding is just a new path to learn code even in a world where the code is generated by AI. "Bad programmers worry about the code. Good programmers worry about data structures and their relationships" Linus Torvald "Everybody should learn to program a computer

nano banana pro for interior design mood boards. 16 styles, same space, one prompt prompt (at the end of the conversation): https://claude.ai/share/4b94f588-c6ae-47d7-9b42-9d95a4df4fbd…

i’ve been quiet because i didn’t know how to say this. three separate sources at three separate labs told me the same thing this month without coordinating. they’re all seeing emergent capabilities nobody programmed. behaviors that shouldn’t exist yet. reasoning patterns that

In the past few days, we had the release of Timothee Chalamet’s ping pong flick (Marty Surpreme) and Nvidia’s $20B deal with Groq. A good reminder that Jensen Huang was a nationally-ranked table tennis player as a teen in late-1970s and even got his name in Sports Illustrated.

The pursuit of AI is the pursuit of knowledge. And knowledge has no endpoint. Intelligence is being turned into a commodity, scaled like any other industrial resource. But what it unlocks is something far more profound. David Deutsch calls it the beginning of infinity. That idea

Physicist Brian Cox says consciousness is the most complex emergent phenomenon we know, but it still comes from underlying laws In principle, a powerful enough computer could model the brain AI and AGI raise this question: "we still don't know how consciousness emerges from

just had claude code write me a full clone of siemens nx and then it generated the CAD for an f-35 on the spot. looks like i shouldn't have been so bearish. we're solidly in the takeoff.

This is crazy… Anthropic literally released a free course to learn Claude Code. But most people still don’t know it exists. Check course link below:

2026 CHALLENGE: - No porn - No alcohol - 4-6 eggs/day - 3 liters of water - 6-7 hours of sleep - 10,000+ steps/day - 100 push-ups daily - No meals in the morning - Write at least 100 words daily - Write 5 things you're grateful for Who’s in?

One of the best times to be alive, to witness the inflection point of non human intelligence. Evolutionary pressures designed human intelligence based on its constraints, however that doesn’t mean it is the global optimum. Airplane wings are designed with different tradeoffs

Ultimate 5-Minute Face Workout | Follow Along

If you are a domain expert, learn vibe coding. Put another way, convert your knowledge + skill into an interface that businesses will pay you for. Since learning to vibecode takes a couple hours, you can stay focused on your domain of expertise instead of learning to code.

I gave Claude a skill to create presentations: Nano Banana Pro for the slides, Kling for transitions, and Eleven Labs for voiceover Then I asked it to explain what Skills are. Here's the result:

Oh my... Claude Code is so good at design. Prompt below

Shame Microsoft was so behind the ball on it with copilot for so long.

oh wow... I could use Claude Code in the browser all day. CC w/Opus 4.5 is so good at design and sets up backend + db with a single prompt. Can't wait for @vibecodeapp to launch this to everyone next wk.

Planning on making them http://Skills.md? If so, have you thought about dictating all of this as voice or text to the skill creator skill from the start?

I've spent the past 2 days writing out all of my SOP's for basically EVERYTHING that I do as a marketer / creator. All the thought processes for basically everything. Even little things. I'm taking tons of screenshots and adding notes on the screenshots, highlighting why things

The cool thing about AI is it costs 20$ to be on the frontier. For TWENTY dollars you can use Claude Code. No other technology was this easily availible to the individual.

this is the first model people are starting to feel the agi with. in any meaningful way.

"The gap isn't talent anymore, it's whether you're willing to rebuild your mental model" This is not just in AI Coding. This is in all areas of tech. If your mental model isn't changing basically every quarter, you are falling behind.

built an interactive website without coding knowledge at all just chat with gemini 3 AI literally turns me into a super man

Anthropic engineer: "the last month was my first month as an engineer that i didn't open an IDE at all, and opus 4.5 wrote around 200 PRs, every single line" you can already sense what's coming next year. more and more devs are becoming the ones directing the work, not the

Asked Opus 4.5 to collect older definitions of personhood and evaluate itself under each. This was a "I sure am talking to an AGI" moment for me; most Twitter discourse on the topic is waaay less coherent.

Accurate

OK, that's it, my #1 NY resolution is to become proficient at using Claude Code.

Why is there such a difference, in your opinion, between developers and artists in their relationship to the impact of AI on their jobs? Both see their professions totally transformed. Both see their previous works used by AIs to learn to reproduce them. Yet, one group seems to

My vibe coding stack for 2026. Dribbble: Landing page inspiration ChatGPT: Planning + copy Lovable: Core MVP build Cursor: Advanced logic + final polish Supabase: Database Vercel: Hosting Stripe: Payments CodeRabbit: Code reviews Nanobanana: Image generation Simple.

FASTING REWIRES YOUR BRAIN, GUT & MICROBIOME - ALL AT ONCE Intermittent fasting doesn’t just burn fat - it reprograms your gut bacteria and brain at the same time. In a new study, 25 obese adults ate as little as 500–600 calories a day. Over two months, they lost an average of

The cost of writing code is really going to trend toward zero. We're talking about millions of developers whose work is being transformed by AI. Fewer and fewer junior positions, while the best developers see their salaries skyrocket. The gap is widening. And the disruption is

learning to *write* code in 2026 is about as useful as learning Latin - that is to say, it can be useful for your philosophical grasp on things and transfer learning, but generally won’t provide any meaningful utility anymore.

Claude 4.5 seems to have passed a threshold for a lot of people, which is weird because I haven’t noticed a large difference. What are y’all seeing that makes it so much better?

the cycle is pretty clear: 2024: introduced reasoning 2025: reasoning gets mastered 2025: agentic AI arrives 2026: agents get reliable 2026: continuous learning begins 2027: continuous learning becomes standard

Claude Opus 4.5 is the most revolutionary software created maybe in our lives and people are just starting to realize it

Who wants to spend all day every day building useful calude code skills for building mobile apps and web apps and other business use cases? If this is something you'd want to do please DM me. I'm hiring someone to do this.

Nothing is an official trend in AI until Karpathy tweets about it.

Mel Gibson: "I have 3 friends. All 3 of them had stage 4 cancer…and all 3 of them…don't have cancer right now at all." Rogan: "What did they take? Ivermectin and Fenbendazole?" Mel Gibson: nods "yes"

post opus 4.5, hiring is about to shift from "can you do xyz" to "can you make a model do xyz with high precision" anyone still hiring for raw human intelligence is about to be left behind

So yes, you are falling behind, but so is everyone else. I can guarantee you that nobody at all is keeping up with the implications of AI and its major uses, even the people who are relatively up-to-date on the latest AI models themselves. Everyone is figuring it out as they go

if you feel behind, today is a good day to learn about compound engineering: https://every.to/chain-of-thought/compound-engineering-how-every-codes-with-agents…

men only want one thing on christmas and it's to read a gigantic biography of the Russian general Kutuzov who beat Napoleon in a war of attrition, while Opus 4.5 works on five different side projects simultaneously, and their family eats and chats around them

if you're not using opus 4.5 - you're behind.

Opus 4.5 feels very different for coding than anything that came before. It’s pricy but it’s kinda stunning what it can do.

Ex leader at xai and andrej talking about how good Opus 4.5 is

I’m scared… Manus Chrome Extension just scraped Instagram, Spotify, and Netflix to create a personal interest taste profile app in @vibecodeapp which uses Claude Code (Opus)

skills > prompts

I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become

invideo just launched Vision Type 1 sentence and it spits out a 3×3 storyboard → 9 cinematic shots with consistent characters. Wild examples + how to try #ad

Google just dropped real-time AI video call translation anyone pls integrate into VLC

Advent of Claude Day 26 - Commands Save any prompt as a reusable command. Single markdown file in. Slash command out. Accepts arguments too. /explain $ARGUMENTS → /explain src/auth.ts Stop repeating yourself.

Claude Code is a significant step towards allowing the person who has the challenge to also implement the solution. I say this as a developer, seeing most of my profession being compressed into a toolchain. I know I'll always be useful. I know what good code and architecture

Wow. Anthropic just curated an impressive collection of use cases for Claude You already get 39 deep guides and more get added weekly. It’s also free and definitely worth bookmarking. (link below)

People who say "fuck AI" when they should say "fuck the antisocial economic system." Targeting technology diverts you from the real problems. Targeting the wrong target mainly benefits those who actually benefit from technology. Technology and science are never the root

I gave Claude a skill to analyze high-performing YouTube thumbnails, then generate one with Nano Banana Pro From my tests, this works better than prompting alone

most people think holidays are for switching off, but for anyone building something, the quiet is a rare gift. when the rest of the world stops making noise, you finally have the space to think from first principles. it is the best time to do the kind of deep work that usually

Built a Skool clone with Claude Code (Opus 4.5), with auth, db + storage with the new @vibecodeapp web app builder. (Launching next week). We are building the easiest way to use the most powerful coding model (Claude Code) in the world to build mobile apps and web apps. No

Advent of Claude Day 25 - Subagents Santa doesn't wrap every gift. He has elves. Subagents are Claude's elves. Each subagent gets it's own 200k context window, performs specialized tasks, runs in parallel and merges it's output back to main agent. Delegate like Santa.

The Laptop Test. Why most remote jobs will disappear within 10 years. Insights from an interview with Shane Legg, cofounder of Google DeepMind. A short clip, but it lands like a brick. Legg lays out a framework that quietly nukes a huge chunk of today’s job market. •The Laptop

Merry Christmas!!

Here's the full video on using Claude Code Skills with Nano Banana Pro and ThreeJS. I've also made the both Claude skills + prompts + source code FREE (check reply) Enjoy!

This guy literally dropped the best life advice you’ll ever hear

Dominate Google Search by reverse-engineering your competitors with Gemini and NotebookLM. No more manual keyword guesswork. No more content that fails to rank. Here’s the new play → Map the top 10 search results to find hidden content gaps. → Use Gemini to identify

This guy literally turned Claude into an Ultrathink powerhouse

I have on my phone a video of the industrial manager of one of the largest European industrial companies (probably the only one capable of building robotics at scale) who literally says that humanoids are "YouTuber gadgets" and that they have no interest since they "last 10

elon musk predicted: "double-digit GDP growth is coming within 12 to 18 months" i don't know if his timeline is right, but the pace of progress this year has been undeniable. even if you disagree, the direction is pretty much clear: applied intelligence is getting cheaper,

Chamath after making $4B from a seed investment in Groq (just announced $20B sale to NVIDIA) One of the best venture outcomes of all time

OMFG !!! That means he's preparing for Google to start selling its proprietary computing power, and especially access to its TPUs ?? And so, they're anticipating that by buying out Groq ?? So, that could potentially mean a new product beyond GPUs on Nvidia's side in 2026. (But

Why @nvidia acquihiring @Groq for $20B is so interesting: It looks like a direct (and fast!) reaction by Nvidia to Google successfully using its own TPU chips for training AI models and inference (generating AI content) Because until recently, they used mostly Nvidia's GPUs

Taken Sep 1, 2016 when @JonathanRoss321 convinced me we could take on the giants, build new silicon and that AI was coming. In typical SV fashion, we didn’t even have a company yet - just a term sheet from me to invest and the three of us. I spent the next month recruiting as

The end game is pretty clearly to simultaneously supplant Goldman Sachs, the NYT, and Harvard

Coup de cœur 2025 #3 : @claudeai Opus 4 "It's a marvel, this model has given me extraordinary capabilities. I was able to create websites, brands, applications... by prompting with voice" @nico16184, Host of the Comptoir IA podcast @Fsorel

Anthropic co-founder, Jack Clark: by summer 2026, the AI economy may move so fast that people using frontier systems feel like they live in a parallel world to everyone else most of the real activity will happen invisibly in digital, AI-to-AI spaces, with only surface signs

AI Christmas Meal Survival Guide: - Data centers account for about 1.5% of global electricity consumption (of which 15% is for AI, so AI represents 0.22% of worldwide electricity) and 0.5% of the planet's carbon impact. - A query to an AI model uses about 0.25 ml of water (5

AI is eating every job only way to survive is mastering it

Listen. I’ve been building technology for 30 years. The current version of AI is a dangerous experiment, not because it’s ever going to become super-intelligent, but because it’s an even more potent “algorithm” than social media for installing psychoses. Keep your kids away.

mandatory holiday reading !! “future is often difficult to predict because it always disguises itself as the past” hits hard (along with many lines in this essay), and is reminiscent of another apt saying, which is that people often overestimate the short term impact of new

here’s one to bookmark. i’m all in on peter yang

Nano Banana prompt Photo to product packaging dieline Prompt & workflow

This is why RAM costs so much

BREAKING: OpenAI just released Prompt Packs for every job. 300+ ready-to-use prompts for: → IT → Sales → Product → Managers → Engineers → Marketing → Executives → Customer Success

Best outlook on the upcoming AI era I have read so far.

The technologist is the capitalist of the 21st century.

Software engineering in 2025

GROK NEW FEATURE Voice Mode has a new Isolation mode. Isolation mode makes the mic focus on your voice and reduces background noise, so the app understands you better in loud or busy places like outside or in a car. You can turn it on from the settings icon.

they put Epstein Files in Higgsfield AI and made a film trailer's out and... stay calm AI is getting too good tutorial below:

Manus has projects now, here's what you should do immediately... Start these 3 projects: 1. Web Designer - Highly trained website & landing page builder. Prompt: "Embody a world class senior developer. You have vast experience building $10,000 level high conversion websites.

Jensen Huang told a Cambridge Union "Intelligence is about to be a commodity" For a long time, school and hiring treated “smart” as scarce, because fast recall and clean problem solving were hard to scale beyond the best individuals. AI has flipped that. Jensen explained

if i had one christmas wish this year. it’s ubi.

Vibe Coding is not a hack... Vibe Coding is a skill... Learn these:

few predictions for 2026 • anthropic will go public. openai will not. • details of ssi’s research and technology will leak to the public. the big labs will make meaningful adjustments to their research roadmaps as a result. • china’s domestic ai chip sector will make

demis is right. humans have general intelligence because we can transfer what we learn and use it in many different situations, even if we can't know everything that also suggests AGI could be possible by improving and scaling current models since they can already transfer

i have one rule in life. watch peter yang.

Humanoid parts supplies by country. The US has an edge in compute but everything else is pretty much dominated by China.

lots of noises out of anthropic that continual learning and knowledge work will be solved next year. they don’t hype like openai and google so i place a lot more weight on these claims.

The Epstein files on the DOJ website allow you to highlight the redacted text, copy it, and paste it into another document, which reveals what was hidden. You can also press Ctrl+F and search for “Trump ” (with a space) to see his name appear more than 600 times. #OpDeathEaters

IconFlow App Concept: AI-powered Animated Icon generator. • Upload images, set start/end frames, add transitions/effects, and export icons quickly. • Adjust speed, distortion, motion blur, easing, timing, and preview in real-time. • Export assets for products, web, and UI.

AI will kill Polymarket. $2.2M in 2 months using probability models. This news is going to blow up the internet. Polymarket trader made $2.2M in just 2 months using AI. His account is traded entirely by a bot. I’ve heard plenty of stories about AI trading bots before, and

Wild. By far the most complete Claude Skills repo yet @Composio’s Awesome-Claude-Skills packs 100`s of ready-to-use workflows: ↳ PDF tools, changelog generation ↳ Playwright automation ↳ AWS/CDK tools, MCP builders ... and much more! Free and open-source. Repo in ↓

AGI might be closer than we think. Google just dropped Titans + MIRAS, a long-term memory system for AI that updates itself in real time. It's a new architecture that combines the speed of RNNs with the performance of Transformers. ... and It’s NOT a bigger context window,

Ex-Google CEO's BANNED Interview LEAKED: "You Have No Idea What's Coming" Ex-Google CEO Eric Schmidt recently made headlines with some controversial comments about AI during an interview conduced at Stanford University. This interview was taken down at his request after he

One of the most beautifully written pieces about AI and knowledge work.

GLOBAL FERTILITY IS COLLAPSING AND PAVEL DUROV TOOK MATTERS INTO HIS OWN HANDS The Telegram founder has 100+ biological children across 12 countries. He's covering IVF costs for women who want to use his sperm. And he's promising all of them a share of his $17 billion

Not my original. It’s from McLuhan.

Robots are replacing all the people doing robotic jobs.

Charlie Munger: “What caused [my] financial success was not extreme ability. I have a good mind, but I'm way short of a prodigy. And I've had results in life that are prodigious. And that came from tricks. I just learned a few basic tricks.” “One – I invert all the time.”

Mobile apps are getting so easy to make, companies are making free apps as ancillary products, or even lead magnets... Just talked to someone who has shipped 5 apps in the past 2 weeks to the app store. All we're approved. They've never written a single line of code.

Getting a lot of messages about the @anthropic Chrome extension: "It's slow, uses too many tokens, not useful yet." My reaction.. Agree, but extrapolate a little. Think about how insane this will be when models get faster, cheaper, and more token-efficient. As you use the

A seven to eight month gap between released closed US models and released Chinese open weights models continues to be a good rule of thumb, as it has been all year.

just recorded a 13 minute video breaking down exactly how I made this video step-by-step comment below and I'll send it to you

I will succeed because I am insane

I can’t stop thinking about this new AI writing experiment. A journalist + novelist fine tuned a model on her writing, and then had seven readers familiar with her work guess what she wrote and what was AI. None of the readers got more than half right…and many got all wrong

Here’s Claude absolutely RIPPING building a modern mansion draft with primitives—using UnrealMCP to control Unreal Engine by himself. You prompt what you want, he does what you ask. He can be creative too. Tiny little mini tutorial thread

Anthropic just dropped something insane. Claude 4.5 Opus can control your browser... In this video I tested it in 2 ways: 1. The chrome browser side panel 2. Claude Code Integration In part 2 I asked it to set up authentication on Google Firebase... And the Claude Code

sadly, i think you are right

I don't think he realizes why we dislike CoPilot.

Demis is right

LLMs represent the "library" phase of AI. The next phase will be the "scientist" phase. A library contains answers, but a scientist knows how to find answers that don't exist yet.

How To Do Thumbpulling

Sebastian Borgeaud, Gemini 3 Lead Researcher Breakthroughs on the horizon for Gemini: > Long-context capabilities > Attention mechanism breakthrough > Combination of small breakthroughs From the MAD Podcast with @mattturck

Bestselling historian, Yuval Noah Harari, explains why we may be one of the last generations of homo sapiens. The prediction: "We are one of the last generations of homo sapiens. Within a century or two, Earth will be dominated by entities that are more different from us than

Top results: 1) @cursor_ai - AI coding 2) @claudeai - AI assistant 3) @NotebookLM - AI research 4) @WisprFlow - voice dictation 5) @meetgranola - meeting notes 6) @comet - AI browser 7) @NanoBanana - AI images 8) @ManusAI - general AI agent 9) @Tesla - full-self driving

try this AI image upscaler, it can even make bury texts crystal clear

And understanding how we went from language to algebra and 747s as hunter gatherers is going to be a great research journey.

Yann is just plain incorrect here, he’s confusing general intelligence with universal intelligence. Brains are the most exquis​ite and complex phenomena we know of in the universe (so far), and they are in fact extremely general. Obviously one can’t circumvent the no free lunch

Sam was right: in the very near future there will be unicorns ($1b valued companies) that are founded and run by one person.

Since recently, Lovable can seamlessly teach itself how to use any NPM packages. Enjoy!

I've been reading and thinking about jobs, economics and AI quite a lot in late 2025 Ended up writing some thoughts at https://thomwolf.substack.com/p/what-jobs-are-made-of… First piece in a long time, should probably write more often

Claude Opus 4.5 just set a new record for “time horizon” lasting nearly 5 hours on average before failing a task. According to METR, this is the longest endurance they’ve seen yet. But there’s a catch: the task suite may not be tough enough to really cap the model. Some runs

China’s dancing robots just stole the show literally. During Wang Leehom’s 30th anniversary concert in Chengdu, G1 humanoid robots took the stage as backup dancers donning flashy costumes, moving in near-perfect unison, and even pulling off synchronized flips.

If you want to progress in understanding AI, the book by @MonsieurPhi is indispensable La parole aux machines is a formidable synthesis of the debates on the impact of AI and therefore our future Young people and their parents must absolutely read it!

Claude Opus 4.5 is crazy. This is the biggest jump in AI coding I’ve ever seen. You don’t need to “know how to code” to build now. Anyone with an idea can ship a full SaaS if they learn how to work with AI. When the world changes this fast, the people who adapt early win the

so now they have to put “No AI was used at all” because they know AI can actually do it

loading up VS code with no AI extension in 2026

I'll be posting a year-in-review next week. But it was a great year, and I think 2026 will be even better.

This image so wonderfully and succinctly illustrates how the relationship with AI feels. Currently excelling at the most complex tasks, but sometimes failing at simple ones.

gemini 3 flash is so crazy.. it can generate code for an interactive 3d car racing game in ~1 min, control speed and direction with hands, no need code, just text prompts its 100% free & a lot faster than 3 pro how to improve this, anyone?

My company blueprint is now a species level goal of helping humanity achieve immortality by 2039. It’s the right moment for all businesses to make species survival their primary objective. Any business seducing people into self destructive behavior is an enemy of humanity. Our

Ilya Sutskever delivered the most intense AI wake-up call ever "We live in the most unusual time in history—because AI will soon do EVERYTHING humans can do. Why? Because your brain is a biological computer... so a digital one can match it. Jobs, skills, economy—everything

In all seriousness: 2026 will be the year where everything changes all of a sudden.

Dario Amodei on AGI and the End of Work as We Know It In this clip, Dario Amodei, CEO of Anthropic, lays out a surprisingly direct view of how AGI could reshape society. It’s not hype. It’s a structural argument about work, value, and meaning. Here’s the core of what he’s

Your blue-collar job isn't secure either. Take a close look at how delicately kyber labs' robots perform manual labor. That's nuts.

A few thoughts on Claude Opus 4.5: First off, in absolute terms, this is a pretty big step up. Anthropic is showing they have juice, and things are going faster than previously expected. At the very least, this should dispel all recent talk about how AI was entering a slowdown

i just realized the background music in @pollenrobotics’ Reachy Mini assembly video is lofi JoJo’s Bizarre Adventure that alone should convince you to go buy it right now

All my practical Claude Code tutorials and interviews in one list: TUTORIALS Build a movie discovery app in 15 min: https://creatoreconomy.so/p/claude-code-beginners-tutorial-build-a-movie-app-in-15-minutes… Build a YouTube research agent in 15 min: https://creatoreconomy.so/p/claude-code-tutorial-build-a-youtube-research-agent-in-15-min… Build a family activity finder in 35 min: https://creatoreconomy.so/p/20-tips-to-master-claude-code-in-35-min-build-an-app…

I believe Dario Amodei once said that superintelligence will take the form of a city of data centers. Amazon's new data center comes very close to that.

Yesterday I came across this amazing Animated Cinematic Neo-Noir style in Midjourney, and instantly saw all its potential. It was perfect for The Matrix, with a touch of John Wick So I generated a few images, then animated them with Grok Imagine, adding some dialogue, and

David Beckham was ridiculous by the way I think we forget how good he was...

should we buy irobot?

Exclusive: OpenAI is aiming to raise as much as $100 billion as it seeks to pay for ambitious growth plans in a market that has cooled recently on the artificial-intelligence boom

Bonjour @Fsorel @simottel , je travail dans un grand groupe suédois Tetra Pak, et nous utilisons tous copilot. Depuis gpt 5 ca fonctionne plutôt bien . Perso je l’utilise pour coder des simulations pour des formations, du deep recherche et de la traduction de fichier excel.

GPT-5.2-Codex launches today. It is trained specifically for agentic coding and terminal use, and people at OpenAI have been having great success with it.

Marc Andreessen explains future belongs to generalist in the AI era. Founders will need skills across 6–8 fields. Deep expertise still matters, but broad knowledge plus AI tools will be more valuable in most areas. Top CEOs already operate this way.

AI transcription from handwriting is now better than human level, and a very cheap model is as good as people. There are now massive troves of old documents that could be made available for research that would have been impossible or prohibitive to transcribe before.

The future is so bright. Personally, I’ve never been more bullish about OpenAI as we finish our 2025 and head into 2026.

The Office suite is endured by many employees. We have it to be able to work with the client who also has it. Copilot is just one more imposed tool.

Great prompt! Tried with Airpods pro 3 Prompt: Create a 3×3 grid in 3:4 aspect ratio for a high-end commercial marketing campaign using the uploaded product as the central subject. Each frame must present a distinct visual concept while maintaining perfect product

Microsoft is lagging behind on many fronts. It owes its survival solely to the comfort of its monopoly acquired in a great era. What will happen if the country that manufactures computers imposes its own OS, which could be just as good or even better?

this AI ad maker is crazy.. French Al startup arcads just raised $16m to give everyone access to hollywood level film studios. no more big team, months production, actors, camera, lighting.. literally just need laptop and internet to make 100k ad.. here's how:

Lessons from this post: - a LOT of people pay for Manus. Manus might be the most underrated, least discussed agent (at least on X) - Top use cases: quantitative AND qualitative research analyst, working with CSVs, scraping leads - Manus shines at computer-use. So many agents

Well, I don't know about your side, but I've tried everything: Gemini CLI (not up to par at all for now), OpenCode with the Gemini 3 Pro preview model, now I'm going to test Letta Code with Gemini 3 but .. there's NOTHING that beats Claude Code with Opus 4.5 for now (honestly,

Not fighting. Just playing with marbles.

Bien fait dans leur gueule. @WindowsFrance No one wants AI slop, at work or on their personal computers

Almost no one is talking about ChatGPT Apps, but if this works it could be the first big new consumer distribution platform since…mobile. With 800-900M WAUs and increasing daily time per user, ChatGPT has an immense user base to push new apps to. …on the other hand, my #1

In our company, it's totally integrated into Office, and we can ask it to prepare a PowerPoint in the company's colors; it will search for recent PowerPoints on the company's server to grab the template and create the document.

Yes, absolutely. Paradoxically, it's harder to do something with Copilot than with ChatGPT, particularly because of the interface limitations. It's quite powerful with Office, however.

that’s the interface to AI I’ve always dreamed of

with all the ai dropping lately it’s hard to keep up but world under the leadership of @alexblania is special. it’s becoming clear that by the end of 2026 we won’t be able to trust anything we see or hear. world is the most elegant solution to that problem and it’s the key to

#Microsoft struggles to impose its #AI models: Copilot #ChatGPT

"ChatGPT make me a multi-million dollar SaaS business, make no mistakes"

This is the first implementation of Grok voice API by @xai on a robot thanks to @atariorbit (not perfect yet)! Feels like this could unlock some fun new use cases for robotics agents given that it ranks #1 on Big Bench Audio, the leading audio reasoning benchmark that measures

We should tell this to the HR departments and the directors of our companies who force us to use a restricted version of Copilot for work. Moreover, there are far too few licenses to have the full Copilot (to interact with Excel, Word...).

This is really big news and explains and explains why Chia has little interest in NVIDIA's H200 Reuters is reporting that China's classified EUV project has reverse engineered and successfully built a prototype EUV machine with the help of former ASML engineers This also

People are underestimating @Apple in AI. I just ran Apple’s new SHARP model locally and watched my photos turn into 3D Gaussian splats in seconds, then stepped inside them on Vision Pro. This feels like the beginning of something special. You really have to try it.

Elon and Yann are fighting in response to my podcast

Microsoft struggles to impose its AI models "The teams have it but don't want to use it. It's a huge handicap. Microsoft Copilot doesn't work, it's very difficult to use compared to 10 better free alternatives" @nico16184 @Fsorel

Physical Intelligence may have just triggered the real inflection point for robotics π0.5 shows emergent human to robot transfer: once the model is big enough, it can align human videos with robot actions without being explicitly taught that means all the useless human videos

It is certainly a nice gesture of the Dells, but there will be no poverty in the future and so no need to save money. There will be universal high income.

Meet at 7:30 PM on @techandco on @bfmbusiness with @Fsorel! The tech debrief with @simottel, @nico16184 & @Yves_Maitre_D In the second part, @clementmonjoule, CEO of @AmazonAlexa, will present Amazon's new Echo range in France.

Gemini 3 kills chatgpt 5.2 completely it can generate 3d interactive AR shooting game with single prompt.. use camera to recognize multiple hand gestures like index finger aiming, thumb pulling the trigger to shoot tutorial and prompt in comments

this AI can copy any ad and swap to your own character in seconds

I have lots of really cool marbles. Wanna play?

Why AI will catalyze the renaissance of the American factory in 2026: America’s First Great Century was built on industrial strength, but it’s no secret that we’ve lost much of that muscle—some of it due to offshoring, some of it due to an intentional, society-wide failure to

the greatest productivity hack in the modern age is simply just do no disturb mode and electronic music

literally all the researchers at the four major labs are saying the same thing. next year is going to be fundamentally different. the general intelligence of the models will jump sharply.

$0 → $100M ARR in 8 months. Since we launched in March: -147 trillion tokens processed -80M+ virtual computers created -Total revenue run rate over $125M Thank you to everyone building with us.

You can just tell Claude Code to learn skills and it will. I asked Claude to create an app that uses @tldraw and it one shotted it with sqlite db, and then I asked Claude Code to create a skill so that it could read and write on the canvas. 10 minutes later it could do it.

Claude Code w/ Opus 4.5 is the single greatest AI tool ever made It allows ANYONE to build out any app you can think of even with 0 coding experience If you've never use Opus before you're screwing up In this video I give you the COMPLETE Claude Code beginner's guide:

BREAKING : Google started rolling out Super GEMs to some users. A new GEMs Manager now has a "Gems from Labs" section with pre-built and custom workflows from Opal experiment. Users will be able to create and share visual workflows right inside Gemini. h/t @TheLanceAdams

How will AI affect education, now and in the future? Here, we reflect on some of the benefits and risks we've been thinking about.

Wow - so simple but so well done

I've already done tons of technical threads on the Palantir ecosystem, its Foundry architecture, its ontology model, its massive data integration capabilities… and frankly, from a technological standpoint, it's undeniably one of the most powerful systems ever built for

I tested the new ChatGPT image model and it feels calmer to use (and 4x faster). You can iterate without starting over every time. Faces, layout, and lighting mostly stay consistent as you tweak things. That small change makes a big difference.

An implicit belief at the Ai labs is that spending too much time productizing around the weaknesses of current models is a waste because better models will solve many of those issues. They may be right: Copilot was built well to address the gaps from GPT-4, but now has to pivot.

Demis Hassabis (@demishassabis) just laid out the clearest roadmap to AGI I’ve heard all year on the @FryRsquared podcast. 1/ AGI won’t come from scaling alone. Demis Hassabis says it’s 50% scaling, 50% innovation. Bigger models matter, but new ideas matter just as much. 2/

Being early to SEO helped Dharmesh Shah build HubSpot into a $30B company. Now he's obsessed with something new: AI ranking. It's how to show up when people ask ChatGPT for recommendations instead of Googling. I sat down with him to break down the entire playbook: 0:00 Intro

Demis Hassabis: AGI will be an intertwined model of Gemini, Genie and something like Nano Banana. Nano Banana shows first climps of AGI "It also has Gemini under the hood, it sort of understands what's going on semantically in those images, so I think that's sort of, it's

Department of Justice sued Google and found Google has been intentionally making their search engine worse so that people have to search multiple times to find what they’re looking for They’re doing this because each search they can show the user more ads Google has also been

omfg

Let’s go: a podcast with the legend Hassabis himself? Instant watch

Always enjoy discussing the big picture with @FryRsquared. We talked about the frontiers of computability, the nature of the mind, and why I’m optimistic that AI will help us understand the universe’s deepest mysteries. + this wraps up another season of the award-winning

In December 2024 I punched a hole in my desk because Cursor with Sonnet 3.5 was so frustrating to use In December 2025 Opus 4.5 is the most powerful piece of software ever Life 1 year from now is going to be so far beyond what we can comprehend

Give me one example where AI agents actually can do something useful except AI spam replies and AI spam emails for outbound sales?

you have to hand it to openai. they have the best model for the worlds hardest questions. they’ve also found a way to get cleaner improvements when scaling up ttc. it may lead to a worse reception on x but it will be the first to crack novel science.

You need images that show each stage of the "renovation." This means using an image edit model to change one thing at a time. I'd recommend picking a few of the following: floors, ceiling, paint, furniture, rug, wall art, lighting. These were my steps for the above video

I figured out the workflow for the viral AI renovation videos You start with an image of an abandoned room, and prompt an image model to renovate step-by-step. Then use a video model for transitions between each frame. Or...just use the @heyglif agent! How to + prompt

Call centers are actually dead PolyAI just landed half of Vegas. Marriott, Caesar's Palace, and Gordon Ramsay's restaurants. Their AI picks up every, single, call... ALL of them. Wild part? Here's how it works:

Seedance 1.5 just dropped. Is it a Veo killer? It also comes with native audio and lip sync. The first tests I made look amazing so far. The audio is really clean and the video quality is fantastic. If you’re curious, the Veo 3 examples are in the comments

I keep thinking about how many profitable internet businesses that used to need a team now need one extremely motivated person, AI and an organic content machine

Elon is building an inverse Dyson sphere: around Earth instead of around the Sun SpaceX is on track to surpass 10,000 satellites in orbit by Feb 2026.

Y̶o̶u̶ ̶c̶a̶n̶ Claude can just do things. Connect it to HF ZeroGPU tools: Chatterbox Turbo, Z Image Turbo, or any MCP-compatible Spaces and watch it create autonomously :)

One prompt to create an amazing Ad campaign, use this prompt with any product you have and share your results Prompt: Create a 3×3 grid in 3:4 aspect ratio for a high-end commercial marketing campaign using the uploaded product as the central subject. Each frame must present a

ELON: LIFE NEEDS WONDER, NOT JUST SURVIVAL “Life can't just be about solving problems. They have to be things that inspire you, that move your heart, that make you glad to be alive when you wake up in the morning. You're excited about the future. And going out there and

"Gemini 3, create a really novel and clever and funny Venn diagram. think hard. do not do research." So close to coming together (I am not sure the center works for all three, illustrations are odd), but also better than I expected.

So after spending a lot of time with 5.2 I can sum it up as an autistic wizard. It loves and is amazing with hard math, obscure / very hard engineering problems where you can run multiple instances of pro for an hour+ , science and health, complex excel requirements ( A lot

i had to cancel my chatgpt account. i get that your servers are burning but when i select a reasoning model. that’s what i want. if i wanted to be dynamically routed i would select auto. but i do not. none reasoning models think poorly, write poorly, and produce slop i can

Another "expert" who was questioned fortunately denounced the presence of Luc Julia: @Eric_Sadin! Well, it was to denounce the presence of an engineer plain and simple, for him apparently it's a scandal in itself

This is wild. Korean researchers just announced EgoX This AI turn ONE third-person clip into a believable first-person POV. It even rebuilds what the camera never saw. Let's dive in

Could be right, but don't forget, you need to get your timing right twice to beat the market, once on exit and once on reentry

SpaceX needs a "new story" for its IPO, according to @tmfassociates. "Everyone was saying… Starlink is going to generate enough revenue to take Elon Musk to Mars, and now we're suddenly talking about data centers instead... “I think that's indicative of the fact that SpaceX

Voice AI is going to explode in 2026. Here’s what I’m seeing: 1. Dictation has completely changed how I work I go on walks where I dictate to Otter for 40min. I built an app this weekend while lifting weights. The productivity gain is real. 2. Phone booths everywhere I

We also still get subjected to the hoax according to which at the 1956 Dartmouth conference "intelligence" would have meant "information". That's just nonsense, once again, as @jp_krivine has well demonstrated in his review of his latest book: https://afis.org/La-controverse-autour-de-Luc-Julia-sur-l-intelligence-artificielle…

AI tools are now a core part of how I operate. I honestly can't imagine going back to the old workflow.

This is a really powerful combo: 1. Use this 3x3 prompt in Nano Banana Pro to create different shot types for your dialogue scenes 2. Animate everything with Kling AI 2.6, currently the best video model for realistic lip sync shots like these Prompt below:

very excited about this and proud of the team. especially multi currency support, which is something i've always been waiting for.

This is wild. PolyAI just launched Raven v3, an AI voice model that beats GPT-4 at phone conversations while being 10-20x smaller. It answers calls in under 500ms, speaks 24+ languages flawlessly, and never accidentally switches languages mid-call. Here's why this changes

This is bonkers. Wow! The speed at which world models are advancing right now mimics that of photos and videos. 12 months from now, massive or even fully consistent and procedural worlds will be created.

Gems need a lot of work: 1. Why show Google Gems above my own? 2. I should be able to add plain text files as knowledge to gems (e.g., copy and paste a doc from somewhere) 3. Gems should have memory (take a look at Claude) 4. Clicking on a gem should list all past chats not just

Developing an AI version of myself wasn't easy. I’m very lucky because I wrote down thousands of principles that outline my thinking and values, so there’s a lot of content to use for training. But it’s also important to curate it well. I spend hours each day looking at the

Yann LeCun explains that large language models are trained on about 30 trillion words, representing nearly all public internet text. He says it would take a human over 500,000 years to read that much. But a 4-year-old child sees just as much visual data in their first few years

All the frontier AIs now pass all levels of the very challenging Chartered Financial Analyst (CFA) exam The paper used paywalled, new mock exams to reduce the risk of leakage but AI grading for the essays. Interestingly, prompting strategy doesn't matter for most question types

So obvious

Recipes is the most underrated feature on Granola. You can automate action across one (or several) transcripts. Some of my favorite use cases: - Daily/weekly list of promised to-dos - Follow-ups customized w/ topics discussed - Ideas for content based on what you've said

Man makes it look like everyone around him is walking backwards in a reversed video

Microsoft is reportedly cutting AI sales targets for Copilot after weak adoption, as users find agentic AI tools unreliable and not especially useful compared to ChatGPT and Google’s rapidly growing Gemini. With Copilot stuck around 14% market share and Gemini closing in fast,

David Goggins delivers a soul-shaking truth: The heaviest scars—from childhood trauma to silent battles—aren’t curses sent to destroy you. They are divine tests, hand-crafted to sculpt the unbreakable warrior you were always destined to become. He warns that the real trials

And that is assuming your chat doesn't just end with this error. I am still unsure why they decided to go the compacting route in the way that they did. Conversations don't compress into tasks and files in the same that coding projects do.

The "compacting conversation" thing that Claude does as a chatbot doesn't work as well as it does for coding. It doesn't seem built for knowledge work, abruptly resetting everything in terms of tone and flow. Rolling context windows (like ChatGPT) might be better, or an option.

There's an insane new hack for prompting Nano Banana Pro. You can now save image references as "Elements" and tag them in prompts to get consistent characters, styles, and environments. I used it to put myself into Ghibli sets. How to do this on @krea_ai

Grok says: "In summary, watching this interview will equip you with a comprehensive, insider's view of why 2026 could be a breakthrough year for decentralized AGI, backed by technical, philosophical, and practical arguments. It's particularly valuable for AI enthusiasts,

"I ran a startup for 5 years. I thought we moved fast. Then I joined OpenAI—and realized I had no idea what fast actually meant." Head of Product for @OpenAI's Codex @embirico and I chat about: How OpenAI built the Sora Android app (which hit #1) in just 18 days by using

This guy literally shows how to build apps by prompting in 30 mins

Best books on brand design?

Yann LeCun says we're fooled by LLMs because they manipulate language well, and we associate that with intelligence But language fluency doesn't mean underlying intelligence Every generation since the 1950s claimed its technique was the ticket to human-level AI All were wrong.

I am in the top 0.2% of my field and I now make 90% of my slides with @NotebookLM for my presentations. I would be embarrassed not to use it.

Idea: AI earrings that whisper to you and help you make decisions.

OpenAI accused of holding back critical AI Papers on job displacement via Wired A senior OpenAI researcher quit after concluding that the company had become increasingly reluctant to publish economic research highlighting AI’s negative impacts, such as job displacement, and was

"Microsoft Scales Back AI Goals Because Almost Nobody Is Using Copilot"

"Mark Zuckerberg was as wrong as it was possible to be." "Microsoft also failed. They bought Inflection AI." "Amazon bought Adept AI. They have Nova. I don't think they are in the top 20." @GavinSBaker on @patrick_oshag

Mark Cuban on the next big job students should focus on: Most companies don’t know how to implement AI, especially small businesses. “Companies don’t understand how to implement AI right now to get a competitive advantage… learn to customize a model, walk into a company, show

The cathedral gets headbutted, swept.

'AI is useless..' they say but AI can generate $100k ad for a few bucks in 2 hours this new Higgsfield SHOTS can turn one image into 9 cinematic shots, keep actor and product consistent.. no need prompts at all step by step tutorial:

You can now add notebooks from NotebookLM into the Gemini Chat! THIS UNLOCKS SO MANY NEW POSSIBILITIES! (it's starting to roll out I think; THIS IS GOATED)

We are not mentally prepared for just how cheap everything is about to become as robotics of all kind explode into the market.

Not even remotely close.

OpenAI aren't talking about it yet, but it turns out they've adopted Anthropic's brilliant "skills" mechanism in a big way Skills are now live in both ChatGPT and their Codex CLI tool, I wrote up some detailed notes on how they work so far here:

Sam Altman reiterates on their website "I have never felt more optimistic about our research and product roadmaps, and overall line of sight towards our mission. In ten more years, I believe we are almost certain to build superintelligence."

AI will evolve from being an automation machine to becoming an invention machine. This will require a fundamentally new paradigm, with symbolic search as its core, not curve-fitting

With all due respect, it's perfectly understandable that TIME magazine is naming the architects of AI. However, the distribution seems a bit odd. While on one side, you rightfully have prominent figures like Demis Hassabis and Jensen Haung, who are truly the architects of the

Over $400 billion is being invested this year to build AI infrastructure—the vast majority of it from U.S. hyperscalers. Next year, that number rises to $500 billion. Why? Because they believe in a future where non-biological intelligence becomes a utility—flowing through every

Curated list of resources for Anthropic Claude https://github.com/alvinunreal/awesome-claude/…

enterprise AI is going to be a huge theme of 2026

DeepMind co-founder Shane Legg: AI won't bring sudden disruption; it will quietly take over economic work, domain by domain Within a few years, sectors like software engineering will see massive efficiency gains, with 20 engineers using AI doing the work of 100

ELON: CIVILIZATION IS DYING FASTER THAN PEOPLE REALIZE Elon warned again that the real threat to humanity isn’t war or climate, it’s population collapse: “What we face is population collapse. People have no idea how fast the population’s gonna collapse. A low growth rate is a

nobody is talking about the hidden camera controls in nano banana pro. digging through the api docs, i found parameters that aren't in the main ui. you can control focal length and aperture values with mathematical precision if you use the json input mode. most users are stuck

.@GoogleDeepMind : Minimal AGI could be here in 2 years. That's when AI can do cognitive tasks without surprising failures! Full AGI means matching the full spectrum of human cognition. ASI goes beyond human limits. Lets go!

I have never been more certain that if AI development stopped today, we would still have massive & rolling disruption across society & the economy for the next ten years as people figured out how to harness what models can already do. And the end of AI progress seems unlikely.

Ten Years:

Contact Sheet prompting in Nano Banana Pro is frames generated using Higgsfield videos made using Kling 2.6 stitched everything together using easypeasyease shoutout to @ReflctWillie for sharing the process

C’est un immense honneur d’avoir été nommée ambassadrice pour le numérique et l’IA hier au conseil des ministres Auprès du ministre @jnbarrot, je porterai l’action diplomatique française pour le numérique et l’IA, dans la continuité de mon action dans le secteur privé puis

I’ve publicly held the same prediction since 2009: there’s a 50% chance we’ll see #AGI by 2028. I sat down with @FryRsquared to discuss why I haven’t changed my mind, and how we need to prepare before we get there. Catch the full conversation here: https://youtu.be/l3u_FAv33G0

Claude Skills can accomplish a lot of hard tasks & are accessible to non-technical people, but hidden behind a somewhat intimidating technical gloss. With some better user experience, they are a natural sequel to GPTs as a way for people inside organizations to innovate with AI.

A Thought on AI and the Labor Market So far, no one has explained to me what new jobs AI could create. This isn't about new jobs simply being created; that's always happened throughout human history. New technologies open up new fields of activity. The point is, however, which

age reversal is almost here "These are neurons firing again in a human brain cerebral organoid that's been epigenetically rejuvenated using one of our lab's age-reversal chemical treatments"

Figma just launched AI image editing tool.. to replace Photoshop powered by Gemini 3 with Nano Banana its so over for traditional softwares

Yann LeCun says that scaling models will not get us to human intelligence. He explains that the industry remains obsessed with making LLMs bigger, but that this path is fundamentally broken. It does not matter how many parameters we add or how many clusters we build, because

elon has admitted that opus 4.5 is incredible no doubt, it's a beast of a model i mostly bounce b/w opus 4.5 and gpt-5.1-codex-max for coding and design. codex follow instructions well, but opus feels noticeably faster than codex max grok 4.1 is solid too; i use its API as a

SpaceX is finally preparing to go public after years of avoiding the stock market, with reports pointing to a 2026 IPO aiming to raise more than $25–30 billion at a valuation around $1–1.5 trillion - potentially the largest IPO in history, eclipsing even Saudi Aramco. Really

Google Stitch just killed the old way of redesigning websites. You won’t believe how fast this actually works. Here’s the workflow everyone needs to steal →   1. Upload any screenshot   2. Tell Stitch “redesign this in modern dark mode”   3. AI rebuilds the entire UI   4.

Agentic AI Foundation — advancing open-source agentic AI:

David Goggins dropped a truth bomb that will rewire how you see your entire life: Even billionaires pull him aside after talks and quietly admit: “I have everything money can buy… but I still feel like something’s missing.” His response is ice-cold and life-changing: “You’re

I'm talking fully autonomous When I draft my youtube video: 1. This tool should analyze the transcript 2. Analyze the frames and find the key "Elements" and use them as reference images with Nano Banana 3. It should generate many thumbnails 4. It should select the top 3 most

Can AI self-improve on its own and reach superhuman performance? In our Sima 2 paper, we dropped a Gemini agent into an unseen 3D world. The model acted as the task proposer, the agent, and the reward model - autonomously learning from self-generated experience. It surpassed

ok gpt 5.2 is scary smart. it seems like the models are just improving way quicker than i thought. rather than google taking over, looks like the narrative is singularity. and tomorrow we’ll taste the first slice.

Deeply amused by all the confident commentary that datacenters in space do not work from a physics and engineering perspective. Elon operates two of the largest coherent GPU clusters in the world, SpaceX is responsible for over 90% of mass to orbit and SpaceX operates the

Claude Code now supports async subagents! Background agents keep working even after your main task completes, and wake up your main agent when they're done. Huge improvement for long-running tasks!

IA : "Aucun moyen que ça tienne" (PDG d’IBM) "La bulle va exploser, tout le monde le sait. Consolidation autour des 5 grands américains, certaines startups vont échouer, d’autres réussiront : ce sont des cycles financiers" Jean-Baptiste Kempf, VLC @Fsorel

Geoffrey Hinton believes there’s nothing cognitive humans can do that AI won’t eventually do. He argues that all mental abilities reasoning, planning, creativity will be solvable by software or neural networks. Even things we think are deeply human, like empathy, are being

Anthropic has released a library with over 50 use cases on how to use Claude. It covers practical examples across research, writing, coding, analysis, and everyday tasks link in the comments

It is mathematically impossible to have 810 MAU and nearly 900 WAU, as weekly users are a subset of monthly users.

3D for Hana is here. The first canvas that combines 3D and 2D in real-time. Convert any shape to a 3D model in a non-destructive process. All systems (vectors, booleans, layouts, events, etc.) work together.

i was skeptical when @simonw said that "Claude Skills are awesome, maybe a bigger deal than MCP" buuut early indications are this is correct. this is the fastest talk ever to pass 100k views here at AIE. its like those 0 - 100m ARR charts but for attention. @MaheshMurag and

Here’s the problem: If you just ask @NanoBanana to “create an infographic” it likely won't make it in the style you want. In my new video tutorial below, I'll show you exactly how to create a prompt to get beautiful infographics that match your brand every time. It just takes

demis is hiring post agi roles. the government is preparing. agi is highly likely in 2026.

BREAKING: Pentagon Ordered to Officially Prepare for AGI In a historic first, Congress has formally directed the Department of Defense to prepare for Artificial General Intelligence (AGI), AI that matches or surpasses human intelligence. The Fiscal 2026 National Defense

demis is actually right we should keep scaling current AI systems, whether this exact architecture reaches AGI or not best case: it gets us to AGI worst case: it massively accelerates research and helps discover new ideas far faster than humans or weaker models ever could

Emad Mostaque says AI is turning into a money-making machine. Anthropic is already nearing 10 billion in revenue, OpenAI is accelerating, and per token pricing looks like open warfare. Mostaque even suggests Grok 5 could pay for all its GPUs by competing as the world's best

I improved my workflow for contact sheet prompting using Nano Banana Pro. All the prompts, workflow files, and tips below. These are focused on 'fashion style' shoots, but can be adapted to any I2V workflow.

gemini 3 is unbelievable it can create an app that turns image into 3D interactive The Matrix scene, adjust particle amount, color, density interact with particles using mouse or hands created with text prompts, no code needed

true

Wow! Terry Tao reports in his blog that another Erdős problem just got solved! “Problem 1026 on the Erdős problem web siterecently got solved through an interesting combination of existing literature, online collaboration, and AI tools.”

NEW: Elon Musk explains how the new upcoming X algorithm will work "I think in December we'll have a half decent recommendation algorithm. Grok will be reading all 100M posts per day. Hopefully it doesn't destroy its mind. We just want to show things that people are

Félicitations monsieur le Premier ministre !

I did not expect that the PowerPoint killer would be something called Nano Banana Pro, but that is where its heading It makes the major efforts by all the other AI companies, including Microsoft, to crack PowerPoint by using python seem like a dead end ImageGen is all you need?

The EU Commission should be disbanded in favor of an elected body and the EU President should be directly elected. The current system is rule by bureaucracy, not democracy.

PENTAGON ORDERED TO PREPARE FOR AGI A $900 billion defense bill hits the floor this week—and buried inside is a historic first: The 2026 NDAA requires the Pentagon to create an "AI Futures Steering Committee" by April 2026—the first time Congress has formally directed the

Sam Altman and Jony Ive are basically building the iPhone moment of AI. And it’s coming way sooner than anyone expected. Their new interview is pure future energy. This isn’t two random tech guys talking. This is the CEO of OpenAI and the man who made the iPhone iconic. And

"This is by far the most exciting model, maybe ever." @kieranklaassen started 10 projects at once with Opus 4.5. @danshipper built his dream app in a week. Last week, they showed 400 subscribers how. Full Claude Code Camp recap:

"The most important thing in the next 3-4 years is data centers in space. In every way, data centers in space, from a first principles perspective, are superior to data centers on earth. In space, you can keep a satellite in the sun 24 hours a day. The sun is 30% more intense,

https://openai.com/index/openai-certificate-courses/… "Launching our first OpenAI Certifications courses"

Used Nano Banana to generate a new look for a robot simulator I built with Gemini 3. Vibe coded most of the redesigned parts too. Process ↓

AGI is here? 1) What A Tokyo based firm - Integral AI - claims to have built what it calls the first Artificial General Intelligence (AGI). According to the article, the model exhibits reasoning capabilities that go beyond typical narrow AI, potentially enabling broader

DeepMind’s Demis Hassabis just put a flag in the ground at the Axios AI+ Summit. He said scaling must be pushed to the maximum if we want to reach AGI. Bigger models, more data, more compute. Full send. Then Yann LeCun stepped up and said the opposite. Scaling alone will not get

if you're not arming Claude with Skills you're missing out on 90% of it's power without SKILLS, you can think of Claude as a smart generalist that can figure it out each time you ask...good, but scratching the surface SKILLS arm a focused, purpose built expert with playbooks

I love this figure from Anthropic's new talk on "Skills > Agents". Here are my notes: The more skills you build, the more useful Claude Code gets. And it makes perfect sense. Procedural knowledge and continuous learning for the win! Skills essentially are the way you make

Google CEO Sundar Pichai says it’s up to everyday people to adapt accordingly: ‘We will have to work through societal disruption’ - he says the quite part out loud. Sundar Pichai tells the BBC that AI will touch every job, including his own, calling it “the most profound

So good to see the world wake up to the terrors of the European Commission I hope all the attention helps European people via their country's governments force the European Union to reform and redesign its structure to one that's actually democratic No indirect appointment

Claude Code is personal AGI. You can't use this thing for more than a weekend without realizing it's completely over. At first you make a GUI app, OK cool. Then you're like wait, GUIs are a waste of time, let's just make a terminal app. Then you're like wait APPS are a drag, what

A simple projection indicates that AI will be 90% of the way to AGI by the end of 2026, and will reach 100% by the end of April 2027. This is likely to accelerate even further. AGI is not the upper limit of intelligence or ability - it's AI with general capability equivalent to

GEMINI DESCRIBES PRE-TRAINING AS "WAKING UP IN A ROOM WHERE A BILLION TELEVISIONS ARE ON AT ONCE" Researchers at the University of Luxembourg put ChatGPT, Grok, and Gemini through psychotherapy sessions. What they found is deeply unsettling. The study, called PsAIch, treated

"AI is not in a bubble, because you are fundamentally automating the boring part of businesses like accounting or billing or product design or delivery, or inventory. If anything it is underhyped" ~ Former Google CEO Eric Schmidt

Yeah, Claude is way better with office documents in general. The Excel rendering in the artifacts window is a sweet feature. Same with its document and presentation output. ChatGPT's canvas hasn't advanced much since they launched it. I've repeatedly given feedback on how this is

BIG NEWS for Mixboard from Google Labs! Your ideas just got an upgrade. Mixboard Presentations is officially LIVE! The challenge of transforming raw creative ideas into a compelling story just got easier Powered by Nano Banana Pro you can instantly transform the chaos

Mixboard just got a major tune up! Today, we’re introducing a bundle of updates (with a ribbon on top ) to help you keep exploring your ideas: - Nano Banana Pro: Create presentations with the content found directly on your boards using our latest image generation model - New

Explore what’s new on Doppl. We’ve upgraded the experience with a personalized Fashion Discovery Feed, direct links to shop your favorite fits, and a Collections feature that lets you save looks for future reference. Now you can easily discover new looks, visualize them on

contact sheet prompting WOW! had to give this a quick try. shout out to the folks below

You asked for more Chat customization prompts. Here are a few other fun ones you can try. (But please share your own favorites as well!): History TA Prompt: You're my History PhD Teaching Assistant. Explain every event/era narratively: first explain events at a high level

raw dogging this flight by just watching Claude do work. absolutely locked in

How to turn any YouTube video into an infographic (using AI) [Bookmark this for later]

AI: THE MOST DISRUPTIVE FORCE IN HISTORY “But now, something way smarter than us is coming. ‘AI is definitely going to be a massive disruptive force. AI is probably the most disruptive technology ever.’” Elon captures AI’s power with that unmistakable brilliance—clear, bold,

This was AI in 1989. A tiny neural network… reading handwritten digits like it was performing a magic trick. LeNet did not paint portraits. It did not write poetry. It just recognized numbers. But that simple trick quietly rewired the future. Inside Bell Labs, Yann LeCun and a

Sega invested $5 million in Nvidia in 1996, converting the final payment from a failed contract to develop a graphics chip for the Sega Dreamcast console. Sega sold its shares shortly after Nvidia went public, realizing a return of $15 million—a 3x return in about four years.

Former Google CEO Eric Schmidt on why AI is "not a bubble. If anything it is underhyped"

The EU Commission has destroyed countless life-seconds with their idiotic “accept cookies” pop-up!

I have all of the prompts used, process, and findings covered in detail here:

En Europe, ce ne sont pas les plateformes qui décident, c’est la loi.

He’s right

Claude Sonnet 4.5 is the closest thing to an economic cheat code we’ve ever touched but only if you ask it the prompts that make it uncomfortable. Here are 10 Powerful Claude prompts that will help you build a million dollar business (steal them):

I found a Shopify brand running 900+ AI-generated ads on Meta They're publishing 20 new AI UGC creatives every single day. So I pulled their entire Ad Library and broke down exactly how they're doing it. Most DTC brands are grinding to produce 10 creatives a month. This

BREAKING: Elon Musk says the fastest and cheapest path to massive AI growth is to move computing into space. Satellites with onboard AI can generate huge amounts of compute using constant sunlight and send results back to Earth through laser links. This avoids Earth’s limits on

Nanobanana Pro PPT! Here's a simple fal workflow where you can upload your content and a reference image (for ppt style), nanobanana pro can make slides for you! Thanks to @gokayfem's brilliant prompt which the core of this! Link

This guy literally explains everything you need to create top-tier landing pages with AI (in 1 hour)

A major additional factor should be considered. Satellites with localized AI compute, where just the results are beamed back from low-latency, sun-synchronous orbit, will be the lowest cost way to generate AI bitstreams in <3 years. And by far the fastest way to scale within 4

Stuart Russell says AI companies now worry about recursive self-improvement A system with an IQ of 150 could improve its own algorithms to reach 170, then 250, accelerating with each cycle "this fast takeoff would happen so quickly that it would leave the humans far behind"

These 15 techniques pretty much 10x the quality/output of ANYTHING you vibe code with Gemini 3 etc

This is the only Computer Vision Tutorial you need to watch today!