What happens when you connect IBM watsonx.ai to a living network of neurons? In this walkthrough, we pair watsonx.ai with the Cortical Labs CL1, a biological computer that runs on lab-grown human neurons, to capture neural signals, extract features, and reason over the results with generative AI.

What is the Cortical Labs CL1?

The CL1 is a biological computer developed by Cortical Labs. Instead of relying solely on silicon, it grows living human neurons on a silicon chip and reads their electrical activity in real time. Neurons fire, spikes are recorded, and those signals can be stimulated and shaped, giving researchers a hybrid platform where wetware and software meet.

Where watsonx.ai comes in

watsonx.ai is IBM's enterprise studio for building, tuning, and deploying AI models. In this project it acts as the reasoning layer on top of the CL1's raw neural data. The end-to-end flow looks like this:

1. Configuration: connect to your watsonx.ai project with your own credentials (never commit real API keys).
2. Record activity: read from the CL1 in a short loop and count neural spikes per channel.
3. Extract features: turn the raw spike counts into a compact set of features such as totals, firing rates, and active-channel counts.
4. Authenticate and analyze: send those features to watsonx.ai and let a generative model interpret the neural activity in plain language.

Why it matters

Bridging a living neural network with an enterprise AI platform hints at a new class of hybrid systems, where biological computation handles certain signals and generative AI makes them interpretable and actionable. It is early, experimental work, but it shows how accessible tools like watsonx.ai can lower the barrier to exploring biological computing. Watch the full walkthrough above, and share your thoughts below.