Coffee with Claude

A memoir of four decades with my doctoral advisor—from beers at The Antlers bar to a joint paper that named my life's work, and the parallel paths we've traveled ever since.

Twenty-one years ago, we built parallel systems to catch when sensors drifted or failed. Now the “sensors” are language models that fail by succeeding too well at the wrong task.

From Mary Shelley's Villa Diodati to living neurons on Lake Geneva to my own distributed sensor network, four distinct approaches to creating intelligence converge on a single question: What is our relationship to the minds we make?

After 36 years watching technology waves reshape ecology research, I’ve learned to distinguish signal from noise. The Collective Intelligence Project’s vision for governing transformative technology is intellectually compelling—but will it become infrastructure or remain aspiration?

I discovered that the autonomous ecological monitoring system I’ve been building isn’t safe to deploy. Two new research papers from Meta, OpenAI, Anthropic, and Google DeepMind reveal that prompt injection—the vulnerability I’d been planning to work around—remains fundamentally unsolved. Here’s what that means for the future of citizen science.