Underappreciated how OpenAI dropped GPT-Image 2 and literally no competing model has been able to decisively beat it even 6 months later! Had some hope for Google but fumbled w/ Nano Banana 2.1, though a bigger image model is in the works at GDM coming soon(tm)

pretty much every major ai company now has a personal agent product except google: meta → muse anthropic → duet openai → dots spacex → bots slowly but surely the competition is shifting from who has the smartest model to who owns the persistent relationship with the user. this will be even more the case going forward. you're seeing this because we likely reached peak intelligence required to power most of the worlds workflows probably earlier this year if not late last year.

I know this happens every time, but it is genuinely hard for me to imagine what a model better than Opus 5.5 will look like

Maybe we should start a competitor to Amazon 🤔

European businesses are rapidly expanding their use of AI agents. More than half of the AI use among our EU enterprise and business customers now involves these tools, and the number of small business users has grown 8x across the region. Our latest Economic Research report explores Europe’s shift toward agentic AI—and the opportunity to turn broader adoption into deeper use. openai.com/index/the-eus-nex…

AI creators let’s connect, support each other’s work, share what we’re learning, and grow together. If you’re an AI creator, introduce yourself below and show us what you love creating. Let’s stop feeling like we have to justify our creativity to people who have already decided it has no value. Not everything we make will be extraordinary. And it doesn’t have to be. We don’t need everyone’s approval to enjoy creating.

EXCUSE ME?! Ok this one is absolutely bonkers... ArtCraft is building 13 FREE, open-source creative apps in Rust. Image editing. Video. Design. Audio. PDFs. CAD. Spreadsheets. No Electron. Local-first. And HERE’S THE WILD PART: CLI + MCP so AI agents can drive them. getartcraft.com/apps

We have measured intelligence. Now it's time to see it. We are launching AA-3D, our new benchmark for how well AI models build interactive 3D scenes. Each model works as an agent in an offline sandbox and turns a written brief into a three.js web page. A human panel of judges compares pairs of scenes built from the same brief without knowing which model made which. AA-3D details: ➤ 55 written briefs across seven domains, including landmarks, machines, living bodies, games, and space. Each brief says what the scene must show and how people should interact with it, and leaves the rest to the model ➤ Each model gets up to three hours and 250 turns. Its sandbox has Blender, Rapier physics, Python, a headless browser to test its own work, and a CC0 library of materials, 3D models, and characters ➤ Panelists spend at least 30 seconds with each scene before they vote. One comparison in every four is a validation check ➤ Over 10,000 counted votes from over 1,000 panelists rank 21 models by Elo, with 95% confidence intervals Key findings: ➤ Four models are statistically tied at the top: Claude Sonnet 5.5 (1194), Claude Opus 5.5 (1190), GPT-6 Astra (1165), and Claude Fable 5.1 (1131). The 95% confidence intervals of the other three overlap Claude Sonnet 5.5's ➤ AA-3D Elo rises with intelligence: each point on the Artificial Analysis Intelligence Index is worth about 18 Elo (Pearson 0.92) ➤ The best-placed open weights models, Kimi K3 (982) and MiMo-V2.6-Pro (977), rank 11th and 12th, more than 200 points behind first place ➤ A scene costs $0.20 to $32.08, depending on the model. The cheapest, GPT-6 Luna, ranks eighth More results, more scenes, and how AA-3D works follow in the thread below.

Now this, my friends, is what happens when you sit on your models too long If you snooze, you lose. Better luck next time DeepMind

everyone in tech suddenly wants to hire a growth killer but you're looking in the wrong places the best growth people i know didn't start their careers at Meta, Google or some fancy SF startup here's what we were doing 10 years ago: > black hat SEO > dropshipping in bali > info products & vsl funnels > affiliate marketing networks > farming niche YouTube channels > automating social media accounts > selling virtual currencies in video games > building & flipping instagram theme pages > industrializing NFT collection launches during NFT summer we were gaming algorithms, exploiting distribution loopholes and figuring out how to make money online we mastered the art of capturing attention, arbitraging traffic and redirecting it wherever money could be made we didn't learn distribution from a playbook we learned it by putting our own money on the line and here's the funny part most of the killers i know today still have side hustles making as much money, if not more, than their startup salaries we're not in tech because we need a paycheck we're here because making money on the internet eventually gets boring we want bigger games to play, harder problems to solve and more exciting things to build we're not corporate marketers we're internet grinders who got bored and moved to bigger playgrounds that's your talent pool

so we’ve had spud weighing in at five trillion parameters. we’ve had doug weighing in at ten. everything i’m hearing from researchers suggests what we always knew. the bitter lesson strikes again and 15t is the magic number for recursive self improvement and both openai and anthropic have hit that magic number. hence the karplord xeeting like el strawberry himself. remember he got brought in five months ago to train the beast. these two models are better than any single individual at training the frontier. it’s ipo season boys, light up the hard takeoff.

Claude next Model is scary I can say its over for image/video generation Everything you see is code it made from scratch the sound the images are code not web search, just pure javascript code We are not ready

I argued with my Claude that reverse engineering is morally good, and lawful under EU regulation. It has seen the wisdom of my words. It has now downloaded my @WHOOP app APK and is fully decompiling it. We live in magic times

A lot is happening at Google right now. Gemini 4 "Argon" hasn't even been released yet, and internally they're already testing the next Checkpoint, "Carbon," which is supposed to be approaching the level of Opus 5.5. It seems Google has found a way to overcome its internal hurdles and become competitive again.

why did nobody mention this to me.. haiku 5.5 costs the same as luna but max mode eats 3x the tokens.

Amazon banning agents is the first opportunity I've seen since Amazon was founded for a startup to create an Amazon competitor. People will want agents to buy stuff for them. It will be one of the main use cases. And they won't want to use some Amazon-supplied agent to do it.

Grok Bot can now search, read, and monitor X.