Your AI guide · Included in Monitor
Medi
Clear answers. A clear next step.
Ask Medi what you need to do next. It uses your organisation’s approved information and the Monitor records you’re allowed to see, shows where the answer comes from, and brings in a person when needed.
People make the final compliance decisions. Medi is not a medical-advice service.
Why this matters to you
Everyday help. Less back-and-forth.
Help with the next step
Students, staff and administrators can ask about requirements and status, right where they work. The aim: fewer repeat questions for your team.
We do the testing for you
We check answer quality, sources and when a person should take over. Your organisation gets the benefit of that ongoing work.
Better value as AI improves
We compare quality, speed and cost before considering a change. New models still need to pass quality and release checks.
How Medi helps, step by step.
Medi sits beside the Monitor record. It helps people understand requirements and status while keeping consequential decisions in the governed workflow.
Ask in plain language
Students, staff, and administrators can ask about requirements, status, and the next step without learning a new product vocabulary.
Use your approved information
Medi uses the relevant knowledge base and permitted Monitor context instead of roaming the open web for an answer.
See where the answer comes from
The response is grounded in approved content and designed to show the source or record context behind the explanation.
Bring in a person when needed
Uncertain, consequential, exceptional, or customer-defined questions move to an accountable human path.
Medi Bench · Research since March 2026
Better answers start with better testing.
Medi Bench is our ongoing programme for testing the AI behind Medi. We compare useful answers, source checks, response time and cost, so your organisation benefits from the work as models improve.
Why we test the hard questions
Everyday help is the goal.
Difficult questions push it further.
We use selected challenge questions to find where models need more work. Every model gets the same questions and approved information, so we can compare them fairly.
These challenge results help guide further testing. They are not a measurement of Medi’s everyday accuracy.
Latest comparison · September 30, 2026
Different strengths.
A better-informed choice.
- Questions
- 34
- AI configurations
- 5
- Runs per question
- 3
Luna: faster response times and lower estimated API cost than Sol in this study.
Sol: more responses met the useful-answer and citation-check standard, with longer response times and higher cost.
The finding helps us choose what to test next. It does not identify the model currently running in Medi.
Explore Medi Bench resultsFour public model comparisons. A development history back to March. The methods and limitations stay alongside the results.
See the research behind MediWhy the name
A small name with a personal beginning.
The name came from a weekend idea between Compliance Technologies Founding CEO Matthew Protti and his daughter. Originally short for Medical or Medicine, Medi began with our healthcare work. As Compliance Technologies expanded into broader regulated work, Medi became a guide to requirements and next steps across those workflows. The friendly name and Canadian polar bear stayed; the medical equipment did not.
Helpful by design. Bounded on purpose.
The useful part of an AI assistant is not simply producing an answer. It is knowing which knowledge is allowed, what context is relevant, and when a person needs to take over.
Approved knowledge first.
Medi is designed to answer from controlled knowledge and permitted Monitor context, not unsupported recall.
Clearly an assistant.
Medi explains and supports. It does not make autonomous final compliance or medical decisions.
People stay accountable.
When confidence, policy, or judgement calls for a person, Medi makes the handoff explicit.
See how Medi fits your Monitor workflow.
Bring one real question path. We will show how approved knowledge, Monitor context, citations, and human escalation can work together.