Satya Nadella (@satyanadella) just published one of the sharpest overviews I've read on how AI is reshaping the relationship between buyers and sellers: the "Reverse Information Paradox."
Here's what I have to add:
Ivan Poupyrev@ipoupyrev
English
Ivan Poupyrev
3.1K posts
Ivan Poupyrev
@ipoupyrev
Founder, CEO at @PhysicalAI. Tech leader and executive, interaction designer, scientist. Google, Disney, Sony before. 2019 National Design Award. TED speaker.
San Francisco & Bay Area, USA Katılım Ağustos 2011
327 Takip Edilen5.2K Takipçiler
The Big Boss himself - Mickey Mouse - stopped by to check out Botanicus Interacticus when we were testing it live at Disney World.
Ivan Poupyrev@ipoupyrev
Years ago at Disney Imagineering I built a novel sensor to make almost anything interactive. As a test you can put a single wire in the soil, and touching a leaf could play a note. I called this project Botanicus Interacticus. The physical world is full of signals and hidden behaviors, and we've never learned to read and understand them. That's why we built Archetype AI. Newton, our foundation model, finally lets us listen.
English
@ipoupyrev You didn’t get to.. the root of the problem 😅
English
Years ago at Disney Imagineering I built a novel sensor to make almost anything interactive. As a test you can put a single wire in the soil, and touching a leaf could play a note. I called this project Botanicus Interacticus.
The physical world is full of signals and hidden behaviors, and we've never learned to read and understand them. That's why we built Archetype AI. Newton, our foundation model, finally lets us listen.
English
@cafecryptoart Thank you! And this isn't even the best Botanicus Interacticus demo. The visual augmentation version we did with Studio NAND crew was absolutely wild. I'll post it sometime.
English
@ipoupyrev Saw the Botanicus Interacticus demo years ago and never forgot it. Newton is very impressive
English
@ipoupyrev Plants as interfaces, absolutely love this. Hiking reminds me that nature is the original complex system; we just needed better tools to parse its signals. Newton is elegant work 🌱
English
@ipoupyrev this looks so so fun but okay, ELI5, what are some meaningful use-cases for this?
English
@Black_Star_Labs 💯 we are also biased by vision. most of the data in the world is non visual.
English
Very cool. I think we’ve been trained to think about signals and sensing all wrong. Always high powered or RF frequency tuned. When everything around us is singals. Sensors can be passive and sense the fields, bioelectrics, and photons, fundamental particles themselves even… it’s everywhere. Interacting with everything. We are ourselves are massive water bodied walking antennas that are emitting light. 😊
English
English
@ipoupyrev @andrewarruda How…. Can I do this… please tell me I can in fact do this
English
@mirrorsrmirror Custom built swept capacitive sensing. This is paper: dl.acm.org/doi/10.1145/22…
English
@tetrisgm Awesome! We did deploy the core sensing tech at the Shanghai & Hong Kong Disney parks, but not with plants. The problem wasn't technical. Living plants belonged to Horticulture, while the interactive stuff belonged to Imagineering. No one could decide who owned the combintion 😆
English
@ipoupyrev Unbeknownst to you, when I ran R&D for Angry Birds, we looked at this and considered its gameplay applications!
I did a lot of research on how to do digital gameplay in real-world spaces back then.
English
Archetype started on one conviction: the physical world is already speaking through sensors — we just couldn't understand it at scale.
Newton Agents change that. Proud of this team. 👏 archetypeai.io/agents
Archetype AI@PhysicalAI
You can now turn a video of a task into a standard operating procedure. 👇 Manual Generation is one of five Newton Agents, built to scale operator expertise before it retires. archetypeai.io/agents
English
This is great point. ChatGPT was free and reached millions being distributed over the Internet.
Industrial automation tech, by contrast, can take like a decade because it's distributed at great cost through multiple levels of midwestern guys in polo shirts.
Ivan Poupyrev@ipoupyrev
English
We’ve seen this before. Case in point is the first VR hype wave started when Jaron Lanier and Autodesk unveiled the DataGlove in 1989 and declared June 7 “Virtual Reality Day.” Great marketing, but the tech couldn’t match the hype. VR became a punchline, setting the field back for years, until the same ideas were repackaged as the metaverse with a modern tech stack.
English
@ipoupyrev this resonates a lot! especially the part with local maxima hypes a-la humanoids that might harm the industry more than help at the moment, when most of those 1X-s of the world are gonna go bankrupt
English
@Doctao87 Agreed, it depends on the definition of what a “ChatGPT moment” actually means. However, if we water it down too much without any hard metrics, then it simply loses any meaning, like what happened to AGI.
English
@ipoupyrev I think of the ChatGPT moment as a function of the demand side rather than the supply side. Basically just a “woah” moment.
English
I'll be at the @autonomousevent in SF on July 16, joining a panel about the models behind Physical AI with @davidlinjiahao, Co-Founder, Abundance, @Sammy_Sidhu, CEO, Eventual, and @petewilz, Chief Product Officer, Vantor.
I’m also looking forward to connecting with others leading the charge in the space like @pathak2206, CEO at Skild AI.
If you're there, come find me. I want to hear how Physical AI is landing in your industry.
English
Ivan Poupyrev retweetledi
The Home of Physical AI & robotics is coming soon to San Francisco.
The Midway, July 16th 📍
The event built to serve the next frontier of AI.
Register below 👇
English
I’m happy to see @PhysicalAI included in the Fast Company world models map.
The physical world is already speaking to us, and now AI models can help operators interpret sensor data more efficiently. Newton, Archetype AI’s world model, turns streams of sensor readings into a single understanding of what's happening and what's coming next.
English