A research lab for
the ethics of
human-AI exchange.

Stoa Lab studies how humans and AI communicate, not the capability of AI systems, but the quality, clarity, and ethics of the exchange itself. We build protocols, tools, and infrastructure for that exchange.

The premise is simple: how intelligence communicates is as important as what it can do.

Something is arriving. Not slowly, not hypothetically. The question is not whether AI becomes a presence in the world. The question is whether it arrives safely: honest about what it is, capable of genuine exchange. That question is being answered, or not answered, in the structure of every conversation happening right now.

AXIS is the first protocol. Nine plain-text operators, no installation, no special interface. When they enter an exchange, something measurable changes.

Tested across 1,000+ exchanges with eight independent AI systems. Consistently: fewer turns, less drift, more precise output. See the data →

AI systems are scaling faster than our ability to communicate with them clearly. The habits being established right now become the standard. We are building the grammar that makes that standard honest.

The grammar is available now. The field is open.