Privacy Policy

Who is responsible

Violema, Inc., a Delaware corporation (“we” or “us”), operates PsychDoc.AI and is responsible for the personal information described here. Max Markovtsev handles privacy requests on its behalf.

Until 12 October 2026, Purple Orange LLC operated PsychDoc.AI. Violema, Inc. is now responsible for personal information received before and after that date.

Make a privacy request

Information you send

The request form collects your name, email, organization, role, request type, and any topic or message you provide. An organization is not required for a correction or privacy request. We also record the page you submitted from, a request identifier, the submission time, and a shortened description of your browser.

If you use a waitlist signup, we collect your email, submission time and a shortened browser description. A briefing inquiry does not add you to a mailing list.

You can read the public research without submitting the form. We need a reply address and enough information to handle an inquiry. Do not send patient records, personal health information, passwords or other sensitive records.

Accounts

You do not need an account to read PsychDoc. An account lets you belong to a workspace and see your plan. You sign in with an emailed link, with Google or with Microsoft.

An account stores your email address, your display name, the kind of client you say you are, your country, your workspaces and your role in each, the version and time of each document you accepted, and the last time you used the account. An account also keeps whether you finished setting it up, how you arranged your home and which team workspace you chose on your home. If you pin a synthesis, a claim or a figure, we keep what you pinned and when. When you open a synthesis while signed in, we note when you last opened it, so your home can show what changed since then. If you change your sign-in email, we keep the earlier address with your account so that closing the account also reaches it, and we delete that record when the account closes. If you sign in with Google or Microsoft, the provider sends us your email address and, if it shares one, your name, and our sign-in service keeps what it sends. An account holds no health information, and we do not ask for any.

A team workspace starts as a request. We ask for the organization’s name, legal name, website and country, and what kind of client it is. People on our team review each request and may keep notes on it. If you invite a colleague, we store their email address, the role you offer and the date the invitation expires, and we send it by email. Other members of a workspace can see your name, email address and role in it.

Signing in sets cookies that keep you signed in and remember where to send you afterwards. We use them for the account only. One of them, pd_signed_in, holds only the character 1 so that pages on this site can show that you are signed in. It holds no name and no identifier. It lasts up to 400 days after it was last set. Signing out removes these cookies.

Accounts are kept in the Violema platform database. Supabase runs that database and the sign-in service, in the United States. Resend sends our email, including sign-in links, invitations and account notices. Both process this information for us.

Evidence you send us

If you send us evidence for a synthesis, we keep what you send: the link or citation, the claim it concerns if you choose one, your note, and the synthesis it is for, with your account and the time. People on our team review it, and we record our decision and, when we have them, a note on it and the synthesis version it went into. You see what you sent and our decision on your account, and we email you at your sign-in address when we decide. We do not publish your note or your name. If we add a source you sent, it appears in the synthesis like any other source.

We keep evidence we decided on as the record of what shaped a synthesis, also after you close your account, but no longer linked to it. Your note stays with it, so do not put personal details in it. Evidence we had not decided on is deleted when you close your account.

Identity verification

Reading PsychDoc needs no identity details. If you ask for a team workspace, the academic rate or a commissioned review, we ask who you are and check what we can.

You give us your workplace and role and a work email address. You may add an ORCID iD, a professional licence (type, number, issuing body and country or region) and a LinkedIn link. We do not ask for identity documents, a date of birth, a passport or a national ID number.

We send a six-digit code to your work address to confirm that you can read mail there. We look up your ORCID iD in the public ORCID record, and a United States licence number in the public NPI Registry run by the US Centers for Medicare and Medicaid Services. Each lookup sends the iD or the number from our server and nothing else about you. We keep the name on the record and the result; from the NPI Registry also the specialty, the status and whether the record belongs to a person or an organization. For other licences, a person on our team reads the regulator’s public register. Software does not read LinkedIn: a person on our team opens the link you give us.

The people on our team who review identity requests can see these details, the results and how you sign in. We write each decision and its date to an audit log without the details themselves. When a reviewer writes a note with the decision, we store it, show it to you on your identity page, include it in the decision email and in your data export.

We keep your identity details while your account is open and remove them when it is closed. Verification codes are stored only as a one-way hash and stop working after 15 minutes. An hourly job deletes each stored code, with the address it was sent to, once it is more than a day old, so it is usually gone within 25 hours. The email that carried the code is outside that job: it stays in your mailbox, and our email provider keeps its copy under its own retention.

Published research information

Research pages cite publicly available papers and research metadata, including author names, affiliations and publication identifiers. We use that information to attribute sources and explain the evidence. Contact us about an attribution error or a concern about personal information in a research page.

Corrections and right of reply

Technical information and browser storage

The service processes an IP address to limit abusive form and sign-in requests. Hosting and security systems may also record IP addresses, requested pages, times and browser information in operational logs.

When you change the theme, your browser stores that preference on your device. It stays until you change it or clear site storage. Signing in sets the account cookies described under Accounts. The site application does not use advertising cookies or advertising trackers.

Fonts, publication marks and compound images are served with the site. Violema’s company website, violema.com, has its own privacy policy. Following a link to a paper, PubChem or another website takes you to a service with its own privacy practices.

Visit statistics

We count visits with Umami, an open-source analytics tool that runs on our own server. It sets no cookies and stores nothing on your device. The statistics are not sent to an analytics company, sold or shared.

For each page view it records the page address and title, the address of the page that linked to it (both without their query or fragment), the browser, operating system and device type it reads from your browser’s description, your screen size and language, and an approximate location (country, region and city) looked up from your IP address. Neither the IP address nor the full browser description is stored.

To count returning visits without a cookie, it derives an identifier from your IP address and browser together with a value that changes every month. It does not know your name or email address.

It also counts a few reading actions: which sections of a synthesis come into view, opening a source paper, pressing a citation’s copy button, switching between the reader and professional views, and sending the request form, with the request type only. None of these include anything you type.

If your browser sends a Do Not Track signal, nothing is counted. Visit statistics are kept for up to 24 months: older records are deleted every day, and backup copies of them expire within 8 days after that.

How we use information

We use submissions to answer the request you made, discuss work you asked about, consider an offer to review research, handle a correction, or respond to a privacy request. Technical information helps us run the site, prevent abuse and investigate errors. We use account information to sign you in, manage your workspaces and plan, and send you notices about them.

We do not sell your personal information or share it for targeted advertising. We do not use contact submissions to make automated decisions with legal or similarly significant effects on you.

Where a law requires a legal basis, we rely on taking steps you request before a contract, our legitimate interests in answering correspondence and protecting the service, or compliance with legal obligations, as appropriate. Where we rely on consent, you can withdraw it by contacting us.

Who receives it

The publisher and people authorized to handle your request may access it. Hosting, backup and email providers process information needed to operate the service. Request records are stored in files on our server or in the platform database described under Accounts, depending on when you sent them.

When email notifications are enabled, form details are sent through Resend to the publisher’s inbox. The notification includes the message and contact details, but not the stored browser description. Waitlist submissions do not trigger that notification.

We may disclose information where required by law or where necessary to protect legal rights or investigate abuse. Service providers may process information outside your country; the applicable arrangements depend on the provider and service. Contact us for details relevant to your request.

How long we keep it

Our retention policy is to keep routine inquiries and waitlist submissions for up to 12 months after submission, then delete them. This covers associated exports, backup copies and notification emails under our control.

Records needed for an ongoing engagement, a legal obligation or a specific dispute may be retained longer for that purpose. Operational logs are retained as needed for security and troubleshooting. The theme preference stays until you change it or clear site storage.

A briefing request for a commissioned review is also copied to a review request record, with the same email address, organization and message. While that record is new, being triaged or declined, we keep it for 12 months after its last change. Once it is offered, accepted, in progress or delivered, it is our commercial record and we keep it.

Records in the platform database are removed by a nightly job when their period ends. Files, exports and copies are cleared in a monthly review by a person. You may request earlier deletion; we will explain any lawful reason that prevents it.

We keep account information while your account is open. You can close it in Settings, under Privacy. We complete the closure within 30 days and email you when it is done. A closure removes your sign-in, your name, your email address and any earlier sign-in addresses, your personal workspace, your place in team workspaces, any identity details, your pins, your reading history, your home settings and any evidence you sent that we had not decided on. Evidence we decided on stays as our record, no longer linked to your account. Invitations sent to you and forms you sent us no longer carry your address, including an earlier sign-in address, up to when it stopped being yours, and a review request your form created loses the question it copied while that request is new, being triaged or declined.

We keep the record that you accepted the terms and this policy: the version, the time, a hashed network address and your browser description. We keep a history of actions on the account for 24 months, by account identifier. We keep review requests that are offered, accepted, in progress or delivered as our commercial record. We keep the other answers on a form you sent us, such as organization, role, topic and message, until 12 months after you sent it.

Your choices and requests

Use the privacy-request form to ask what we hold about you, correct it, request deletion or stop follow-up. Include the email address you used and what you want us to do. If you have an account, you can change your display name, client type, country and sign-in email in Settings. We may need to verify that the request concerns your information, through your signed-in account or a reply from the address you used. Do not send identity documents.

Depending on the law that applies, you may also have rights to a copy of your data, portability, restriction, objection, withdrawal of consent or a complaint to your local data protection authority. These rights can have exceptions. We will respond within the applicable legal timeframe.

Make a privacy request

Children and sensitive information

The service is intended for adults researching scientific evidence. We do not knowingly collect information from children under 13. Contact us if a child has submitted information or if sensitive information was sent by mistake so we can address it.

Make a privacy request

Changes to this policy

We will update the date on this page when practices change. If you have an account, you see the new version on your next visit and accept it to keep using your account. If a change requires notice or consent under applicable law, we will provide it. Contact the publisher through the privacy-request form with questions.