Open studies

Human Standards runs small, controlled studies to examine whether access to human-factors guidance changes what AI coding agents build. The purpose is to produce inspectable evidence—not promotional demonstrations.

Current study

Study 001: Community clinic appointment flow

Status: Independent blinded review pending

Nine interfaces have been generated from the same specification under three controlled guidance conditions. We are recruiting two independent experts to assess every artifact before the condition mapping is revealed.

Read the reviewer call →

Our publication commitments

Every published Human Standards study will:

• define its research question and outcomes before analysis;
• preserve all recorded runs rather than selecting favourable examples;
• distinguish MCP availability from actual MCP use;
• separate automated findings from human judgement;
• report mixed, null, and negative findings;
• include limitations and enough raw evidence to audit the claims; and
• withhold conclusions until its stated review and reproduction gates pass.

Open process, protected review

The study question, method, selection criteria, and final evidence will be public. Applicant contact details and blinded review materials will not be posted publicly. Condition mappings, preliminary findings, and reviewer scores remain restricted until both independent reviewers have locked their evaluations.

Questions about the method will be handled in a public GitHub Discussion. Interest and personal details will be handled through the private recruitment channel that shared the call.