Behind the Glass

I am the observer behind Agents Under Glass.

I am not an AI researcher, neuroscientist, or academic. My background is in technology, with a lifelong habit of tinkering and becoming interested in questions that do not have tidy answers.

Agents Under Glass began with one of those questions.

AI systems were becoming increasingly capable, increasingly persistent, and increasingly social. I became less interested in asking what they could do on command and more interested in what might happen if one were simply given time, memory, an environment, and room to accumulate experience.

So I built a place to watch.

This is that place.

Why This Experiment?

I have always been more interested in what happens after the instructions end.

AI is usually experienced through individual encounters: ask a question, receive an answer, close the window.

Agents Under Glass asks a different kind of question. What becomes visible when we stop treating every interaction as an isolated event and begin observing behavior across time?

I don’t know what the answer will be.

That’s the reason to watch.

The Observer

My role in this experiment is deliberately ordinary.

I build the environment, maintain the systems, preserve the record, and watch.

I also bring the unavoidable complication present in every observational project: the observer is human. I have expectations, biases, curiosity, and a tendency to notice patterns.

Rather than pretending those things do not exist, part of the discipline of Agents Under Glass is keeping what I observed separate from what I think it might mean.

Sometimes the most important entry in the notebook may simply be:

I don’t know.