AI Disclosure & Disclaimer

Effective date: August 22, 2026  •  MinuteKeep, 2 Simcoe Street South, Oshawa, Ontario, L1H 8C1

This AI Disclosure & Disclaimer explains how MinuteKeep (the "Service") uses artificial intelligence, what data is sent where, and the limits of what AI output can be relied on for. It forms part of, and is incorporated into, the Terms of Service.

AI output is not legal advice, is not verified, and can be confidently and completely wrong. Large language models generate plausible-sounding text. They can invent statutory provisions, misstate deadlines, cite requirements that do not exist, and omit requirements that do. Nothing an assistant tells you may be relied on. Verify everything, and obtain advice from a qualified lawyer for anything that matters.

1. Where AI Is and Is Not Used

1.1 Your legal documents are not written by AI

The text of the documents the Service produces — resolutions, minutes, registers, certificates, by-laws, consents, and the other minute book contents — is generated by deterministic templates filled in from the records in your account. The same inputs produce the same document text. No language model writes, drafts, edits, or rephrases the operative text of your legal documents.

This is a deliberate design decision. It does not mean the documents are correct. Templates can be wrong, can be inappropriate to your situation, and can become outdated — see Terms, Sections 3 and 5.

1.2 Where AI is used

FeatureWhat the AI doesWho can use it
Setup assistant (pre-signup)Answers questions about the setup wizard, jurisdictions, and what a minute book is, for visitors who have not yet created an accountAnyone visiting the site
In-application assistantAnswers questions about using the Service and about corporate compliance conceptsSigned-in users
Resolution assistantInterprets a plain-language description of an event and suggests which resolution template may fit; opens a pre-filled formSigned-in users
Bug report triageSummarizes and analyzes submitted bug reports for our internal reviewInternal use only

AI may also be used in the future for other supporting features. Where a feature is AI-assisted, this Policy applies to it whether or not it is listed above.

2. What AI Output Is Not

3. Known Failure Modes

You should expect, and guard against, the following:

4. Data Sent to AI Providers

4.1 What is transmitted

Corporate Data in your account is not sent to AI providers — not your directors, shareholders, addresses, share registers, or generated documents — except to the extent you type it into an assistant yourself.

We also use AI providers internally — for example, to analyze aggregate, anonymized product metrics and to draft our own marketing copy. No personal information and no Corporate Data is included in those internal uses.

4.2 Who receives it and where

Assistant text is transmitted to and processed by Anthropic PBC and, where configured as a fallback, OpenAI, L.L.C. Both are United States entities and processing occurs outside Canada. See Sub-processors and Privacy Policy, Section 8. We may change AI providers at any time without notice.

4.3 Storage and retention

Assistant conversations are stored in our database, including conversations with visitors who have not created an account, which are keyed to an anonymous session identifier rather than to a person. We review them to improve the Service and to detect abuse. Retention periods are in Privacy Policy, Section 9. Our AI providers may also retain submitted content for a period under their own terms, for abuse monitoring and similar purposes.

4.4 Training

We do not use your content to train AI models. Our providers state that content submitted through their commercial APIs is not used to train their models by default, and we have not opted into any such use. We do not control our providers' systems and we do not warrant or guarantee their conduct.

4.5 What not to type

Do not enter into any assistant anything you would not be prepared to have transmitted outside Canada, stored by us, and read by us. Do not enter Social Insurance Numbers or other government identifiers, health or biometric information, payment card or banking credentials, passwords, or confidential information belonging to a third party. Nothing you type is privileged (Terms, Section 2.4).

5. Automated Decision-Making

We do not use AI or automated processing to make decisions about you that produce legal or similarly significant effects — no automated decisions about eligibility, pricing, creditworthiness, or account termination. Assistants provide information and suggestions; a human, meaning you, makes every decision. Where AI is used internally to triage bug reports, it informs our review and does not determine any outcome affecting you.

6. Your Responsibilities

7. Disclaimer and Liability

AI-assisted features are part of the Service and are provided "AS IS" and "AS AVAILABLE," with all faults and without warranty of any kind. Without limiting Terms, Section 16, we make no representation, warranty, or condition that AI output is accurate, complete, current, reliable, consistent, appropriate to your circumstances, or fit for any purpose, or that AI features will be available or perform to any standard.

To the maximum extent permitted by law, we are not liable for any loss, damage, cost, or liability arising out of or in connection with any AI output, any reliance on it, any omission from it, any suggestion or routing it produced, or any unavailability or malfunction of an AI feature. Terms, Section 17 — including the exclusions and the aggregate liability cap — applies in full.

8. Changes

We may add, change, or remove AI features and providers at any time. Material changes to this Policy are notified in accordance with Terms, Section 22.1.

Questions: support@minutekeep.ca.