My god this is such a good speech that every SWE needs to hear. You know what? Every person should hear it
Keep the happy memories, eyes on the reality, be excited about the future. That’s the best that anyone can do
I asked Opus 5.5 to explain camera focus by building an interactive lens lab
Here's what it came up with after 1 hour 26 minutes in one shot, $25.66 API cost
https://t.co/uqDgUs8H9h
Move the focus ring and you can see the glass elements shift the sharp plane through the scene
I'm truly impressed with Opus 5.5. I asked it to visualise Atlantis in ThreeJS for me, and it created a tour around Poseidia, the last surviving island of Atlantis.
Sound on 🔊 for full trailer experience. Even this trailer is made by Opus 5.5 as well.
I think we’re on the verge of a new type of media: Generative Experiences.
The idea is that you can describe a world you have in mind and step into it is pretty incredible. You could explore its cities, meet its inhabitants, discover its stories, and shape what happens next through your own choices.
This took Opus 5.5 about two hours to make. But if we extrapolate a year into the future, I can imagine being able to conjure up these kinds of fantastical, immersive experiences within seconds.
The possibilities go well beyond gaming. Writers could walk through the worlds they’re building. Children could explore an ancient civilisation and ask questions along the way. Learning could become an experience you actively participate in, with the world responding to what you’re curious about.
This is incredibly exciting and I am here for it.
You can tour around Poseidia here:
https://t.co/RlIsego8Ze
FYI, I am not sponsored by meta at all.
I just really love cool stuff … and sadly most of the ai products we have right know are seriously uncool. This is because they are built by people who aren’t cool in the first place, it’s just engineers building for other engineers who think the whole world is made of code and everything runs in a terminal.
Apple was the most valuable company on earth for over a decade because of one simple reason. ANYONE could pick up an iPhone and use it and get value out of it, from young kids to grandmothers. You don’t have the equivalent of that today in AI, we just have a bunch of nerds with psychosis and early signs of schizophrenia who would rather talk down to their users and tell them AI is too dangerous and is going to kill everyone… that’s not grandma friendly.
I think people forget that technology and knowledge by extension is the great equalizer. Many will try to gate keep this stuff but I give kudos to meta for actually innovating in this domain and making a product anyone can use while still being fun and friendly!
Muse is 1,000x more valuable and important than just another coding model with a 2% eval bump that everyone will pretend is great and will just be used to write unit tests. The bottleneck was never the model it was the fact that you have no taste, no agency, and for that reason you don’t build anything impactful.
You know what’s really impactful?
An agent that is personalized to me, that gets my jokes, knows my taste, has my swag. A thing that knows me inside and out is worth a hundred times more than a thing that knows everything else but that and always talks down to me. This should be a stark warning to all the AI labs, you risk alienating your users more by patronizing them and limiting their experience. There is only so much coding market share left and what comes after that is just ten times harder and more difficult to monetize instead of just doing good for your users.
Good luck 👍 & shout out Meta for really delivering a game changing experience !
Muse is genuienly really good. I can't believe Meta was first to this type of thing. How did they beat Google, Anthropic, and OpenAI to it?
I'm sure they will all release their version of it too, but right now Muse is way more impressive than I thought it would be.
the most underrated part of muse is the model that powers it. thank you to the MSL research team—some of the most technically brilliant, hardworking scientists and engineers i’ve ever known. you brought original ideas to difficult problems, invented new methods, and battled through enormous technical complexity to build the jewel at the heart of muse.
thank you to the teams working on agentic post-training, tool and computer use, memory, personalization, alignment, safety, and multimodal understanding. thank you to the people building data, evals, and infrastructure beneath it all. so much scientific ingenuity lives in those details: designing a better experiment, finding the flaw in an apparently good result, staying with a stubborn problem until a new way forward becomes clear.
and now that jewel powers a beautiful product your grandma can use without knowing a thing about the extraordinary complexity underneath. it will just work for her. all those discoveries, all that patient work, becoming something simple and useful in another person’s hands. there is something magical about that. i’m grateful to be building with you.
Muse for Mac is out today! It works across apps, files, calendar, notes, and messages on your computer. You control what it can access. The team is shipping fast. Download at https://t.co/BX3oX19Zcl
@meta@Muse has already helped me book a CA DMV appointment for wife yesterday, today it renewed my car registration, scanned my email for open items and create calendar events for others. Crazyy usecases.🔥 always dreamed of having this.
@alexandr_wang need more ideas 😁
I spent 2.5 billion tokens, every limit I had and five days to make this.
A GTA-style city, built with GPT-6 using only prompts. No code or 3D by hand. The whole point was to push it as far as it goes and wire in every API I had.
The setup: Codex driving Unreal Engine 5, with Higgsfield, Tripo, Meshy, Seedance, Seed Audio and fal handling the buildings, characters, animations, materials and sound. 8 buildings taken from concept to game, 247 generated voice lines for residents and police, 10 radio tracks across 5 stations.
Here's the honest part. Most of those tokens didn't go into building. They went into fixing. The model writes a mechanic, runs it, watches it break, and tries again for hours. Getting a character to open a car door and actually sit in the seat took an absurd number of passes. A carjacking needed two separate bodies to stay locked to the same grab. Half the work was animation that looked fine on its own and fell apart the moment it ran in the game.
And almost all of that 2.5B is the same project history being re-read from cache on every request, not unique code. Procedural generation from prompts is genuinely fun to watch and wildly inefficient at the same time. The city started as glowing blocks with working driving, money and jobs underneath, and ended up somewhere I'd actually screenshot.
Full breakdown: https://t.co/pmdiG1uIZU