Mindreading with AI
Understanding the Language of All Living Beings
2026 COIN (Collaborative Innovation Networks) Seminar at University of Cologne, University of Bamberg, HSLU
Mindreading with AI
Understanding the Language of All Living Beings
2026 COIN (Collaborative Innovation Networks) Seminar at University of Cologne, University of Bamberg, HSLU
Everyone talks about flying to Mars. But what if the most important unexplored frontier isn't millions of kilometers away — it's right beneath our feet, in our gardens, in every forest and ocean?
What if AI could help us understand what plants feel, what animals know, what fungi decide?
This isn't science fiction. It's happening now. And you can be part of it.
We're using AI and machine learning to decode communication across all living beings:
• Humans — emotions from voice, face, and text
• Animals — whale songs, bird calls, bee dances
• Plants — bioelectric signals that respond to sound, light, and human emotions
• Fungi — underground networks that share information across forests
• Bacteria — quorum sensing and collective decision-making
The goal: build AI systems that can translate between species — creating a kind of universal translator for all life on Earth.
Mars represents escape — the idea that if we ruin this planet, we can start over somewhere else.
Interspecies communication represents integration — learning to live as part of Earth's living system, not as its exploiter.
"If we can't communicate with the life forms that share our planet, we have no business spreading to others."
This project combines three elements:
• Sensors — capturing bioelectric signals, sounds, chemical signals
• AI/Machine Learning — pattern recognition, classification, prediction
• Human Interfaces — sonification, visualization, haptic feedback
Current results from our plant research:
• 95% accuracy — detecting sounds from plant bioelectric signals
• 85% accuracy — detecting light changes
• 73% accuracy — classifying human emotions nearby
• Apply machine learning to real-world sensor data
• Work with time-series and signal processing
• Use LLMs as coding assistants (the new way to build)
• Design human-computer interfaces for non-traditional data
• Collaborate in high-performance teams using COINs methodology
• Contribute to real research with publication potential
This seminar uses a unique team methodology based on 20+ years of research into how the best innovations actually happen — from Linux to Wikipedia to the Transformer architecture that powers ChatGPT.
You'll form Collaborative Innovation Networks (COINs) — small, self-organizing teams that achieve "groupflow." We use AI-powered analysis of your own team communication to help you collaborate better.
→ See: Creating COINs — The Team Methodology
→ See: Sample Project Ideas
→ See: Cybernetic Alchemy
This is frontier research. Nobody knows all the answers. That's what makes it exciting.
You'll use AI to explore questions that matter: Can we understand what other species experience? Can we build bridges across the communication gap between all living things?
The biggest breakthroughs come from people who try things nobody else is trying.
For more info also read this blog post of Peter about Cybernetic Alchemy
For more information contact the main instructor, Peter Gloor, at pgloor1@uni-koeln.de
Websites of previous years
https://sites.google.com/view/coinseminar25
https://sites.google.com/view/coinseminar24-25
https://sites.google.com/view/coinseminar24
https://sites.google.com/view/coinseminar23-24/
https://sites.google.com/view/coinseminar23
https://sites.google.com/view/coinseminar22-23
Seminar Papers of previous years
coinseminar 25