An Analog Cognition Project

THE LIVING MIND

A Walk-Through Artificial Brain

Watch intelligence come to life.

A room-scale physical neural computer where visitors can stand beneath analog neurons, watch optical signals travel through the network, and experience normally invisible computation unfolding around them.

Explore the vision Help build it

Concept rendering. The project is currently in development.

The vision

Step inside a functioning physical neural network.

There are countless ways to interact with artificial intelligence on a screen. The Living Mind is being developed to create something fundamentally different: a place where visitors can stand beneath a brain-shaped network and watch computation unfold through physical state, electrical thresholds, and pulses of light.

The suspended installation will contain dozens of analog artificial neurons. Each neuron will receive signals, accumulate charge, fire after reaching a threshold, and communicate with other neurons through optical pathways. Even at rest, the network will maintain a gentle baseline pattern of activity.

When a visitor asks a question, the entire experience will awaken. Conversational AI will construct the spoken answer while a separate physical neural network receives real signals, evolves according to its own dynamics, and makes its computation visible throughout the exhibit.

What is really computing?

Three connected layers, clearly explained.

The exhibit will be dramatic, but its scientific credibility depends on showing exactly what each part of the system does.

The language model

A conventional conversational AI interprets the visitor’s question and constructs the words of the spoken or displayed response.

The physical analog network

Real neuron circuits integrate electrical state, fire at physical thresholds, transmit optical signals, and perform selected temporal, memory, and classification tasks.

The visible experience

Every illuminated neuron and pathway corresponds to a real state or transmitted event. The lighting reveals activity rather than inventing it for decoration.

The suspended brain is not presented as a literal life-size view of every parameter inside a large language model. It is an independent physical neural computer working alongside one.

Concept rendering

Central experience

Ask the Living Mind

Visitors speak or type a question at the main station. The network begins from a quiet baseline, then visibly moves through listening, interpretation, physical activation, response construction, speech, and a gradual return to rest.

01Listening
02Interpreting
03Activating the physical network
04Constructing the response
05Speaking
06Returning to baseline

Interactive experiences

Touch the ideas, then watch the brain respond.

Each station is designed around clear physical cause and effect, not just another screen or button-driven animation.

Build a Perceptron

Adjust physical connection weights until a simple learning machine correctly separates patterns into two groups.

Excite or Inhibit

Send positive and negative signals into a neuron and watch its physical state rise, fall, and cross the firing threshold.

Create a Memory

Place a signal into a recurrent loop and watch the pattern continue after the original input has disappeared.

Inside an Artificial Neuron

Control an enlarged transparent neuron and see integration, leakage, threshold, firing, reset, and recovery happen step by step.

Follow a Thought

Select one physical signal and trace its real journey through neurons, optical pathways, and competing activity.

Slow Down Thought

Replay a recorded firing sequence slowly enough to follow every signal and state change across the network.

Additional full-installation experiences

What Happens When Connections Fail?

Disable a pathway, silence a neuron, or add controlled noise and discover whether the network adapts or loses the pattern.

Train the Mind

Reward selected outputs and watch adjustable physical synapses change strength over repeated trials.

Inside a Token

Explore how a language model divides text into tokens and uses context to predict what should come next.

Two Minds, One Stimulus

Compare an educational view of biological processing with The Living Mind’s live physical response to the same input.

Race the Machine

Compete in pattern recognition, reaction time, and prediction challenges that reveal different strengths in people and machines.

Take Home Your Thought Pattern

Receive an anonymous image or animation of the unique firing pattern created by your interaction.

The Path to Intelligence

Eleven stations from programmable machines to physical minds.

A circular timeline surrounds the brain. Visitors can begin anywhere, but the numbered path builds a complete story of artificial intelligence.

01
1830s to 1936
From Calculation to Computation
Charles Babbage, Ada Lovelace, and Alan Turing
Programmable machines began as mechanical designs, then became a universal theory of computation. Before machines could think, people first had to imagine machines that could compute.
02
1943
The First Artificial Neuron
Warren McCulloch and Walter Pitts
A simple mathematical neuron combined several inputs and produced an output according to logical rules. The model established the idea that complex behavior could emerge from networks of small units.
03
1950
Can Machines Think?
Alan Turing
Turing replaced an impossible definition of thought with a practical question about behavior. Could a machine communicate well enough that a person could not reliably distinguish it from a human?
04
1956
Artificial Intelligence Gets a Name
The Dartmouth Workshop
Researchers gave the field a name and explored several paths, including symbolic rules, search, learning systems, and brain-inspired networks.
05
1957 to 1960
A Machine That Learns
Frank Rosenblatt and the Perceptron
The physical Perceptron learned simple visual patterns by adjusting its own connection strengths. It became a direct ancestor of modern neural networks.
06
1969 to early 1990s
Limits, Detours, and AI Winters
Early limitations and changing research priorities
Hardware, data, and methods could not yet support the field’s largest ambitions. Neural research slowed while other approaches gained attention, but the core ideas survived.
07
1982 to 1989
Networks Learn Hidden Representations
Hopfield, Rumelhart, Hinton, Williams, LeCun, and others
Recurrent networks demonstrated physical memory, while backpropagation made it practical to adjust connections across multiple layers and learn useful internal features.
08
2006 to 2012
Deep Learning Breaks Through
Large datasets, graphics processors, and multilayer networks
Better algorithms, more data, and powerful hardware converged. Deep networks began outperforming older methods in image recognition, speech, and other real-world tasks.
09
2017 to present
Attention, Transformers, and Generative AI
Modern foundation models
Transformers used attention to find relationships across language and other data. They became the foundation for systems that converse, create images, write software, and generate complex content.
10
Here and now
The Living Mind
A physical neural computer
Conversational AI and a separate physical analog network work together. The language model constructs the response while the suspended machine performs real, visible computation of its own.
11
The future
What Comes Next?
Physical intelligence beyond conventional computing
Neuromorphic circuits, analog state, optical communication, new materials, and embodied systems may reshape what intelligent machines become and how much energy they require.

Development roadmap

Build the proof first, then build the world around it.

The project advances through testable physical milestones. Each stage must be stable, observable, and maintainable before the next begins.

One Physical Neuron

Prove optical input, analog integration, threshold firing, reset, and visible state.

Four Connected Neurons

Demonstrate excitation, inhibition, recurrence, optical propagation, and a simple learned behavior.

Sixteen-Neuron Prototype

Create the first museum-ready lobe with baseline activity, visitor controls, and live instrumentation.

Flagship Museum Installation

Build the full suspended brain, timeline ring, question station, and complete visitor experience.

A Worldwide Network

Place 20 to 40 Living Mind installations in science museums around the world.

The worldwide vision

Twenty to forty Living Minds, connected across the world.

Each installation can be adapted to its museum and community while sharing a common physical architecture, educational mission, and global identity. Future locations could exchange small anonymous heartbeat signals and occasionally create synchronized pulses across the entire Living Mind Network.

One day, a visitor might see:

27 Living Minds are awake around the world.

Help build The Living Mind

This project needs many kinds of minds.

We are interested in hearing from engineers, researchers, educators, museum teams, fabricators, sponsors, and others who can help turn the concept into a working public installation.

Analog and mixed-signal electronics

Optical and neuromorphic computing

Artificial intelligence and language models

Museum and interactive exhibit design

Structural engineering and fabrication

AI history, neuroscience, and STEM education

Science museum partnerships

Sponsorship and investment

Important questions

What The Living Mind does and does not claim.

Is it a real computer?

That is the goal. The suspended network is being designed around physical analog neuron circuits, adjustable synapses, real state, recurrent activity, and optical communication. The first prototypes must demonstrate these capabilities before the full museum installation is built.

Is the language model physically inside the brain?

No. A conventional language model provides the conversational layer. The suspended brain is a separate physical neural computer that receives real signals and performs observable computation alongside it.

Is it alive or conscious?

The name is intentionally poetic. The network will remain active and respond to its environment, but it is not a biological organism. Visible activity demonstrates computation, not consciousness or subjective experience.

Why build it physically?

Modern AI is usually hidden inside chips, data centers, and software interfaces. The Living Mind makes computation tangible, inspectable, and understandable while also exploring analog and optical approaches to future computing.