Dr. Gotthilf[CloakLLM]For clinics and medical practices
Let your clinicians use AI on real cases. Keep your patients’ identities in the clinic.
CloakLLM removes names, record numbers and contact details before a question reaches an AI model, keeps the clinical detail the answer depends on, and puts the patient’s details back into the reply. It runs on your own systems.
The situation in most clinics
AI is already in your consulting rooms. Usually unofficially.
Doctors use it to summarise a history, check an interaction, or draft a referral letter. It saves real time. But every one of those questions can carry a patient’s identity to a company your clinic has no agreement with.
Banning it rarely works
Staff who save an hour a day find a way. The risk just moves out of sight.
Allowing it unchecked is worse
Patient identifiers in third-party AI services are a data-protection incident waiting to be discovered.
There is a middle path
Let staff use AI, with identities removed before anything leaves the clinic, and a record of what was sent.
What it looks like
The same question, before and after.
A fictitious note, run through CloakLLM set up for clinical text.
Without CloakLLM: what the AI provider receives
Patient Maria Lopez, MRN 00482913, DOB 03/14/1961, phone 555-201-8833, email maria.lopez@example.com. 64-year-old woman with type 2 diabetes, HbA1c 8.9%, eGFR 52.4, on metformin 1000 mg twice daily. Should we add an SGLT2 inhibitor?
With CloakLLM: what the AI provider receives
Patient (removed), MRN (removed), DOB (removed), phone (removed), email (removed). 64-year-old woman with type 2 diabetes, HbA1c 8.9%, eGFR 52.4, on metformin 1000 mg twice daily. Should we add an SGLT2 inhibitor?
Name, record number, date of birth, phone and email stay in the clinic. Age, diagnosis, lab values, medication and the question go through, so the answer is still useful, and the doctor sees it with the patient’s details restored.
What your clinic gets
Safer AI, without slowing anyone down.
Identities stay in the clinic
Patient identifiers are removed on your own systems before any AI request goes out.
Nothing changes for clinicians
They ask their question as usual. The protection happens in between.
A record for your data-protection officer
Every AI request is logged in a tamper-evident record that itself contains no patient data, and can be checked independently.
Open source, no lock-in
Your IT team or vendor can inspect every line. You keep the setup and the records.
How we work with you
Start with a data-protection assessment.
01
Listen
Which AI tools your staff use, what goes into them, and which rules apply to your clinic.
02
Measure
We run CloakLLM with your IT team or vendor, on your systems, on real examples. Patient data never leaves them.
03
Report
What is protected, what is not, and a setup that closes the gaps, measured again.
04
Protect
We help your IT team or vendor put it in front of the AI tools you choose.
A fixed fee for the assessment, quoted after a first conversation.
Plainly
·CloakLLM protects AI tools that your clinic or your vendor controls. A doctor typing into a public chatbot in their own browser is a different problem, and we will tell you honestly what helps there.
·It hides who the patient is, not the clinical detail. That detail still goes to the AI, which is why the provider and the use still need your review.
·It works on text. It does not read scans, images or audio.
·It helps you meet your data-protection duties. It does not make a clinic compliant on its own.
Find out what your clinic’s AI tools are sending about your patients.
Tell us which tools you use. We will reply with a scope and a fixed price.