- Developers
- Learn about building generic AI models for neuromotor interfaces and potential applications.
- Business
- Understand how AI-driven interfaces can enhance product accessibility and user experience.
- Everyday users
- Discover how future technology might allow intuitive control of devices with your thoughts or movements.
This episode offers a solid interview discussing advancements in AI-powered interfaces and their accessibility benefits.
77: How to build a generic neuromotor interface
Audio streams from Meta. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
The Meta Tech Podcast episode "77: How to build a generic neuromotor interface" features Pascal Harris interviewing Jesse, Lauren, and Sean. They discuss how AI enables device control through intention or simple movements, eliminating personalized training. The conversation covers non-invasive interfaces, their advantages over older methods, and their potential for accessibility.
This episode discusses new technology that lets you control computers or devices with your thoughts or simple movements, like flicking your wrist. AI helps make these interfaces work for anyone without needing special setup, making technology easier for everyone to use.
This episode explores the cutting edge of human-computer interaction, focusing on neuromotor interfaces that use AI to interpret intentions for device control. It's for anyone interested in accessibility tech, AI's role in simplifying user interfaces, and the future of how we interact with machines. The discussion on generic interfaces and AI's role aligns with Meta's broader AI research and development efforts, particularly in areas aiming to make technology more intuitive and accessible.
- HostPascal Harris
- GuestJesse
- GuestLauren
- GuestSean
- Episode Title: 77: How to build a generic neuromotor interface
- Show: Meta Tech Podcast
- Length: 31 minutes
- Key Technology: Generic neuromotor interfaces
- AI Role: Eliminates need for personalized training
- Interface Type Discussed: Non-invasive
Join Pascal as he explores the groundbreaking world of generic neuromotor interfaces with Jesse, Lauren, and Sean. Discover how these technologies enable control of devices with just a flick of the wrist or even a simple intention to move. We'll discuss the role of AI in eliminating the need for personalised training, the differences between non-invasive interfaces and their predecessors, and the exciting implications for accessibility. Don't miss this deep dive into the future of human-computer interaction.
Got feedback? Send it to us on Threads (https://threads.net/@metatechpod), Instagram (https://instagram.com/metatechpod) and don't forget to follow our host Pascal (https://mastodon.social/@passy, https://threads.net/@passy_). Fancy working with us? Check out https://www.metacareers.com/.
Timestamps
-
Intro 0:06
-
Jesse introduction 1:29
-
Lauren introduction 2:42
-
Sean introduction 3:29
-
Team's mission statement 3:49
-
What's a neuromotor interface? 4:24
-
Paper overview 5:29
-
Non-invasive interfaces 7:50
-
How to make it generic 9:42
-
Design tradeoffs 11:29
-
Real-world model performance 14:21
-
Feedback cycle 16:22
-
LLMs and EMG 17:22
-
Handwriting vision 18:39
-
Working with product 20:55
-
EMG for accessibility 22:25
-
How Meta helps 25:53
-
Open-source repos 28:02
-
What's next? 28:45
-
Outro 30:51
Links
-
A generic non-invasive neuromotor interface for human-computer interaction - Nature - https://www.nature.com/articles/s41586-025-09255-w
-
How the low-vision community embraced AI smart glasses - The Verge - https://www.theverge.com/the-vergecast/701018/ray-ban-meta-smart-glasses-be-my-eyes-ceo-accessibility-tech
-
MKBHD on Orion - https://www.youtube.com/watch?v=G0eKzU_fV00
- neuromotor interface
- Jesse, Lauren, and Sean
- Technology that lets you control devices with your brain or body signals.
- non-invasive interfaces
- Jesse, Lauren, and Sean
- Interfaces that do not require surgery or penetration into the body.
- EMG
- Jesse, Lauren, and Sean
- Electromyography, a technique to measure muscle electrical activity.
87: Everything is Power: Battery Life Engineering for AI Glasses
WithRey, Dmitrii
86: A Hard Cell: Engineering Ultra-Narrow Batteries for AI Glasses
WithKarthik, Myuran
85: Reel Friends: Building Social Discovery that Scales to Billions
WithJoseph, Subasree
Same guests, other shows
54: Building Key Transparency at WhatsApp
WithSean, Kevin
Questions people ask
- What is a neuromotor interface?
- A neuromotor interface allows control of devices using signals from the brain or nervous system, such as intentions to move or muscle activity.
- How does AI help these interfaces?
- AI eliminates the need for personalized training by enabling generic interfaces that can interpret user intentions without specific calibration for each individual.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Meta.

