1. The Short Version
This summary is for convenience. The numbered sections below are the policy.
- What we collect. Your account details; the code, database results, documents and facts you connect or enter; what the Services produce from them; billing records; and how the Services are used.
- What stays on your Mac. Database checks run on your Mac against a read-only role, and only measured results leave. Work done by a model on your Mac stays on your Mac. When the Mac app uses a hosted model, your request and the code it needs are sent to an AI model provider.
- AI models. We do not train AI models on your content. Cloud features send content to the model providers listed below, which keep it for the time their own terms allow. We do not promise that a provider deletes it at once.
- We do not sell your information or share it for targeted advertising.
- You choose what others see. Banks, processors, law firms and regulators see what you share with them.
- Not legal advice. We are a software company. Information in the Services is not covered by attorney-client privilege.
- Your rights. You can ask to see, correct, export or delete your information. Section 13 says how.
2. Who We Are and Our Roles
For your account, billing, usage and website-visit information, we decide how the information is used. For the code, data, documents and facts that you or your organization put into the Services ("Your Content"), we act on your organization's instructions as its service provider. If your organization needs a data processing agreement, write to us at the address in section 17.
Apparently AI LLC is a software company and not a law firm. Apparently Law PLLC is a separate company that shares an owner with us. This Policy covers Apparently AI LLC only. A law firm handles what you give it under its own engagement letter and professional duties (section 9).
3. What We Collect
Account information. Your name, email address, organization, role and sign-in details. Sign-in is handled by our authentication provider. If you use multi-factor authentication, we store what is needed to verify it.
What you connect and enter. Repositories, websites, documents and opinions you connect or upload, and the facts you enter about your product, company and states. From your database, the Mac app sends us only the results of the checks it runs against a read-only role, not rows of data. If you turn on the database change feed, a collector that you run reads only the tables and columns you allow, updates the same aggregate checks as changes arrive, discards every row at once and uploads only measurements and a digest. Where that is not possible, we poll for the same measurements every few minutes. If you connect a GitHub repository, we read it to test your product, which downloads a copy of the code to our servers.
What the Services produce. Findings, scores, suggested ways forward, drafts, readiness packets, review records and the decisions you and the institutions you work with record, kept as an auditable history.
Billing information. Card details are collected and held by Stripe, our payment processor. We keep your Stripe customer and subscription identifiers, plan, amounts and dates. We do not keep your card number.
Model usage records. For each use of a hosted model: which model, the amount of text processed, the cost, the time and the organization. These records do not hold the text itself. The work the model produces for you is stored in your workspace like any other output.
Usage information. Page views and actions on our website and in the product, recorded in our own database (section 10), and the device, browser and country our servers can see from a request.
Communications. Messages you send us, and email we send you, including delivery status.
Counsel applications. If you apply to review opinions as counsel, the bar admissions you list, the records of how each was verified, and conflict-screening notes entered by our staff. Staff read these records; the applicant does not.
Key-person information. If you use licensing workflows, information a regulator requires about owners and key persons, such as identification and background details, that you or they provide. We hold the most sensitive of these fields in encrypted form.
Opinion checks. Where a payment processor can upload an opinion it received so that it can be checked, we hold that file and the findings for 30 days, as section 11 describes. The uploader is often not the author. The author's information is in the document.
Other sources. Information that a bank, processor, law firm or regulator you work with adds about you in the Services, and public records about companies and licenses.
4. What Stays on Your Mac, and What Leaves It
- Database checks. They run on your Mac. The Mac app shows you what it will send before it sends it. What leaves is the check results and a digest of the mapping. Rows, player names, email addresses, street addresses, IP addresses, the text of the queries and your credentials do not leave your Mac.
- Models on your Mac. The Mac app can run models on your Mac through a local model runner. Those runs do not send your content to an AI model provider. If you choose a model that your local runner hosts in its own cloud, that provider receives your request under its own terms.
- Hosted models. The Mac app's default mode uses your Mac first and the hosted model only when your Mac cannot do the job. You can set it to use only your Mac, or to prefer the cloud. When the hosted model is used, your request and the parts of your code it needs are sent through our account to an AI model provider. A single request can carry a large amount of text.
- Connected sources. Repositories and documents you connect are read by our servers.
You can remove the Mac app's key from your account settings at any time, and the app then stops reaching your account.
5. AI Models
Where content goes. Cloud features send the content a task needs to AI model providers: Anthropic for most tasks, Voyage AI for the numerical representations (embeddings) used for search and similar-document matching, and OpenAI for a small number of functions. The Mac app's hosted model is one of these.
Training. We do not train or fine-tune AI models on Your Content, and we do not give a model provider permission to train on it.
Retention by providers. A provider processes content to return a result and may keep it for a limited time under its own terms, for example to detect abuse. Whether a provider keeps a copy, and for how long, depends on its terms and the settings on our account with it. We do not promise that any provider deletes content at once or keeps nothing.
Privileged matters. When a task involves a matter marked privileged, a filter masks client names and personal details and blocks passages of protected documents before anything goes to a provider, and the task runs on a model on your Mac instead where it can. No filter can promise that every identifying detail is removed.
Accuracy. Model output can be wrong. The Terms of Service say what that means for how you use it.
6. Learning Across Customers
We do not use one customer's content to train a model. These are the ways information from many customers is combined:
- Measurements. We use measurements that cannot be tied back to you, such as risk levels, outcomes and response times, to run and improve the Services, and to show peer comparisons. A comparison is shown only when enough different organizations contribute to it.
- Collective intelligence (opt-in). It is off unless an owner or administrator of your workspace turns it on. If it is on, a model rewrites finalized work product, such as an approved opinion, to remove identifying details, and the generalized version is made available to other customers, including your competitors. You can withdraw consent in settings, and what your organization contributed is withdrawn too.
7. How We Use Information
- To provide the Services: run reviews and checks, score readiness, prepare documents and packets, and deliver what you share to the parties you choose.
- To run your account, bill you and support you.
- To keep the Services secure, prevent abuse and enforce our Terms of Service.
- To improve the Services, using the measurements described in section 6.
- To send service messages and, if you opt in, our newsletter. You can unsubscribe at any time.
- To comply with law and respond to lawful requests.
We do not sell personal information, and we do not share it for targeted advertising. Where the law of your place requires a legal basis for processing, ours are: to perform our contract with you or your organization; our legitimate interests in security, support, measuring use and improving the Services; your consent, for things you opt in to; and legal obligations.
8. Who We Share It With
People you choose. When you share a review, packet or evidence with a bank, payment processor or law firm, or accept an invitation from one, that party sees what you shared. A bank or processor may provide its files to a regulator that supervises it, including through Apparently's regulator workspace.
Service providers (subprocessors). We use providers to run the Services. They may use your information only to provide their service to us. We use:
- Supabase: database, sign-in and file storage;
- Vercel: hosting and scheduled jobs;
- Stripe: payments and billing;
- Resend: email we send;
- Cloudflare: bot protection (Turnstile) on some forms;
- Anthropic, Voyage AI and OpenAI: AI models and embeddings (section 5);
- Google: fonts our pages load (Google receives your IP address and browser details when a page loads them);
- Sentry: error reports, only when error monitoring is turned on;
- Twilio and Slack: text messages we send and internal notifications.
If you choose to sign in with Google, GitHub or LinkedIn, or to connect GitHub, DocuSign, Microsoft (Outlook and Teams), Google (Calendar and Gmail) or HubSpot, those providers receive what that connection needs. For scheduling with law firms we use Calendly, and for background checks that licensing workflows require we use fingerprinting and background-check providers (Fieldprint and IdentoGO). We may add or change providers and will update this list.
Law firms. If you ask a law firm for legal work, we send it the information it needs, and it handles that information under its own engagement and professional obligations. When a bank, payment processor or vendor orders a review of your opinion, attorneys of the firm it engages see the files you shared with it and handle them under that engagement.
Legal and safety. When required by law or legal process, or to protect the rights, property or safety of Apparently, our users or others.
Business transfers. In a merger, acquisition or sale of assets, subject to this Policy.
9. Apparently Law and the Separation Between Us
Apparently Law PLLC is a law firm and a separate company. It receives your information only when you, or an institution that engaged it, share it with the firm. We are a software company: what you put into the Services is not a communication with a lawyer, and attorney-client privilege and a lawyer's duty of confidentiality do not apply to it because of the Services. A law firm's own duties attach when you sign its engagement letter.
Our staff may open Your Content to run, secure and support the Services, as the Terms of Service describe. Staff access to the most sensitive identity fields is logged.
10. Cookies and Analytics
Cookies we need. To sign you in, keep your session secure (including protection against cross-site requests), carry you through sign-in with another provider, and remember settings. Blocking them may stop sign-in from working.
Product analytics. We set a first-party cookie with a random identifier that lasts 90 days. With it, our own server records page views and steps in sign-up and in the product, with the page address (which can include the part after a question mark), the referring site, campaign tags and, when you are signed in, your account and organization. It is kept in our own database and is not shared with an advertiser. You can block or delete the cookie in your browser. We do not currently show a cookie banner, and the cookie does not honor Do Not Track or Global Privacy Control.
Public pages counter. Separately, a counter on our public pages records the page path, the referring site, campaign tags, a country and a device type. It sets no cookie, keeps no IP address or user agent, and turns itself off when your browser sends Do Not Track or Global Privacy Control.
We do not use advertising cookies or third-party ad trackers.
11. Retention and Deletion
- Account information. Kept while your account is active. If you delete your account in settings, your name and email are replaced, your sign-in is removed and you leave your organization. The organization's records stay.
- Workspace content, findings and review records. Kept while the workspace is active and afterward as an audit trail. We have not set an automatic deletion date for them. Evidence shared with a bank, processor, law firm or regulator may also be kept by that party under its own obligations.
- Opinion checks. The uploaded file and its findings are deleted after 30 days by a daily job, or sooner on request. A small record remains: the uploader's email domain, a hashed form of the address, the month, counts and cost.
- Data export. The download link we email expires after 24 hours.
- Billing records. Kept as long as accounting, tax and dispute rules require.
- Model usage records. Kept as part of your billing history.
- Product analytics and error logs. Kept in our own database. We have not set an automatic deletion date for them.
- Backups. Our providers keep backups on their own schedules, so deleted information can remain in a backup until it ages out.
You can ask us to delete information. We will delete or de-identify what we can, and tell you what we must keep and why, such as audit records, records under a legal hold, and records that other parties hold.
12. Security
We protect the Services with encryption in transit; access controls in the database and the application that are meant to keep each organization's data separate; optional multi-factor authentication; hashed storage of API keys; extra encryption for especially sensitive fields such as key-person identity data, database connection details, stored credentials and multi-factor secrets; and audit logs of significant actions. We do not claim a third-party security certification. No system is perfectly secure, and these controls are not a guarantee. If a breach affects your information, we will notify you as the law requires. To report a security problem, write to team@apparently.cc.
13. Your Rights and Choices
Depending on where you live, you may have the right to know what personal information we hold about you and to access it; to correct it; to delete it; to receive a copy in a portable format; to object to or limit certain uses, including sensitive information; to withdraw consent you gave; and not to be treated worse for using these rights. We do not sell personal information or share it for cross-context advertising, so there is nothing to opt out of.
How to ask. Use the export and delete-account tools in your settings, or email team@apparently.cc from the address on your account. The export tool covers your account records and some workspace records; if you want everything we hold about you, ask by email. We may need to verify your identity before we act, and we may ask for more detail if a request is unclear. An authorized agent may ask on your behalf with your written permission.
When we respond. We respond within 30 days. Where the law allows, we may extend that once and will tell you why. There is no charge for a reasonable request.
Organization accounts. If your information is in a workspace run by your employer or another organization, that organization decides how it is used, and we may refer your request to it.
If we say no. We will explain. You may reply to ask us to reconsider, and you may complain to your data-protection authority or state attorney general.
14. International Transfers
Apparently is based in the United States, and the Services are hosted there. If you use them from elsewhere, your information is transferred to and processed in the United States, and by our subprocessors wherever they operate. Where the law requires a transfer mechanism, we rely on appropriate safeguards such as standard contractual clauses. We have not appointed a representative in the European Union or the United Kingdom.
15. Children
The Services are for businesses and are not directed to anyone under 18. We do not knowingly collect information from children. If you think a child has given us information, write to us and we will delete it.
16. Changes to This Policy
We may update this Policy. We will post the new version here with a new date and, for material changes, notify account owners by email or in the app before they take effect. We will not use a change to apply a weaker practice to information we already hold without your consent.
17. Contact
Questions about privacy, or a request under section 13? Contact us at team@apparently.cc. We have not appointed a data protection officer. Questions go to the same address.