Maieutic Intelligence Limited · Hong Kong
Trusted AI, beyond promises.
We build traceable, governable AI work systems that support accountable oversight—from independent assessment and evidence capture to controlled pilots.
01 / Why MAI
AI systems now influence real work, real decisions and real institutions. MAI makes their evidence and controls inspectable—so adoption can move faster without leaving governance behind.
02 / What we build
From the first procurement question to the final human approval, we design for evidence—not theatre.
MAI Second Opinion
AI proposal review
Examine problem fit, vendor claims, cost structure, architecture and governance gaps before procurement, investment or deployment.
MAI GovAI Evidence Layer
AI evidence and governance layer
Create traceable records across models, data, tools, actions, evaluations, human approvals and exceptions.
MAI Applied AI
Applied AI work systems
Co-design bounded human–agent workflows with limited permissions, human approval and measurable outcomes.
Evidence, not a black box
A defensible AI system records more than its final response. It keeps the chain of inputs, sources, tools, evaluations and human decisions available for review.
03 / The maieutic method
A method of bringing clearer answers to light through disciplined questions. For AI: What did it do? What evidence supports it? Who approved the action? Does the result merit scaling?
Define the decision, workflow, risk boundary and success criteria.
Review the data, models, tools, vendor claims and controls.
Run a bounded workflow with limited permissions and human approval.
Measure quality, cost, risk and governance overhead before scaling.
04 / Where we work
Pre-procurement review, validation of vendor claims, decision evidence capture and governance of cross-department AI workflows.
Teaching, research and administrative agents supported by knowledge boundaries, rubric-based evaluation and human review.
AI vendor assessment, controlled automation, permission and approval design, and verification of quality and cost.
麥楚林
05 / Company
Michael's work spans productivity and efficiency measurement, industrial economics, and how AI changes organisations and work. He founded MAI to translate those questions into practical methods for AI assessment, governance and deployment.
06 / Start here
Start with one proposal, one workflow, or one unresolved governance question.
If you are assessing an AI vendor, designing a pilot or building a traceable governance mechanism, let's talk.
Tell us what you need