Home · Technology · Nov 20 archive
Humanoid Robot Iris Runs on Lab-Grown Neurons
· Confirmed
In Short: Researchers have connected lab-grown human neurons to a humanoid robot named Iris, enabling real-time responses to social cues.

Researchers at Stanford University have successfully connected lab-grown human neurons to a humanoid robot named Iris, allowing it to respond to social cues in real time.
The neurons, derived from human skin cells and developed into clusters called organoids, are attached to electrodes and integrated into Iris's control system.
Cameras fitted into Iris capture its surroundings, feeding this information to the neurons through an application programming interface (API). The neural activity then influences Iris's behavior, enabling it to interpret basic social cues and adjust its responses accordingly.
Liphardt explained, “When you and I do anything with our hands or our mouths, we're doing it between a 10 and 100 millisecond loop.” This lag time is similar to how Iris processes information.
The robot's hands are designed for grasping and manipulating objects, while its cameras and sensors help it interpret its surroundings.
Liphardt noted, “If you close your eyes and turn away and ignore her, she will cool.” This indicates that Iris can perceive and react to social cues in real time.
While the concept of biocomputing may seem unusual, it could lead to more efficient and adaptable robots in various applications, from home assistance to industrial use.
The use of lab-grown neurons in humanoid robots represents a significant step towards biological computing, opening new possibilities for real-time interaction.
Robots that are becoming remarkably capable machines at their physical control still struggle with the higher order control required to learn quickly or adapt when conditions change.
The information inside the neurons cannot be easily read out, which matters if it’s been living in your home.
These technologies are at an early stage of development, but this could potentially lead to a future where robots are better at navigating complex environments.
What this adds
This technology could revolutionize the way humanoid robots interact with their environment.
The firm shipped more than 5,500 humanoid robots last year as demand for the technology grows.
Background
Humanoid robots are being deployed in warehouses, homes, and even at marathons, signaling a new era of automation.
What's confirmed
- 13 — remotely connected Iris to two commercial neural platforms, both made of hundreds of thousands of human cerebral cortex neurons: 2D cultures from an Australian company and 3D spheroids from a Swiss company.
- “It’s critical to explore alternative compute architectures,” he said.
- That neural activity then fed back into Iris’s behavior, enabling the robot to interpret basic social cues, such as a person engaging with it, and to adjust its responses accordingly.
What's still developing
- Since the turn of the century, scientists have been exploring ways to convert this vision of hooking up neurons to a robot to reality, with the earliest experiments connecting lamprey brain cells to drive a robot in 2000. 1 One of the people working toward this dream is Jan Liphardt, a bioengineer at Stanford University and founder of the company OpenMind.
- The firm shipped more than 5,500 humanoid robots last year as demand for the technology grows.
- The unveiling of this technology comes amid a growing push to develop humanoid robots capable of performing increasingly complex physical tasks.
- This technology could revolutionize the way humanoid robots interact with their environment.
- Su was more skeptical about the future of such an approach, though he said he does not disregard biological integrations into robots.
