Terms of Service
Effective date: March 30, 2026. For existing users, these Terms take effect on March 30. For users who create an account on or after March 15, or who purchase credits before the effective date, these Terms apply at the time of sign-up or purchase.
Contact: contact@beaverapp.ai
Who we are. Beaver is an AI-assisted research tool that integrates with reference managers and document editors (such as Zotero and Microsoft Word). Features include semantic search, a reading assistant, chat with your library, and automated processing. We are not affiliated with Zotero, Microsoft, or any other third-party platform.
By installing or using Beaver, you agree to these Terms. If you don't agree, please don't use the Service.
1) What the Service does
Beaver stores your library data locally on your device and uses server-side processing for AI features. We access titles and abstracts to generate search embeddings; route your prompts/queries and relevant document context to AI models (whether Beaver‑managed or user‑selected; see Section 3); and provide search. Some features are beta/experimental and may change or break.
Cloud Library Sync. Cloud Library Sync is not enabled by default and currently applies only to early-beta accounts that were onboarded with this configuration. When enabled, your library data (metadata and files like PDFs) is synced to our servers for enhanced features such as advanced search, document processing, and cross-device access. We may offer Cloud Library Sync under additional plans in the future; the list of eligible plans may be updated at any time. Cloud Library Sync is never enabled without your explicit action. See our Privacy Policy for details on data handling under each configuration.
2) Your content & your responsibility
Ownership. You keep all rights to your library, PDFs, notes, and metadata ("User Content").
License to operate. You grant us a limited license to host, copy, and process User Content solely to provide and secure the Service for you (e.g., create indexes/embeddings, generate outputs you request, and provide support). Any product‑improvement use of AI Data occurs only if you opt in as described in Section 4; we do not use your Files for product improvement.
You must have rights. You are solely responsible for ensuring you're legally allowed to use each document in your connected libraries with Beaver—including having it processed, indexed, and sent to AI providers. This applies to copyrighted, licensed, paywalled, or embargoed content. If licenses or institutional permissions are required, you must obtain them. We do not review your materials for legal compliance.
Representations. You represent and warrant that you have all rights and permissions necessary to use your User Content with Beaver (including content under paywalled, licensed, or embargoed access) and that your use complies with applicable licenses, institutional policies, and laws.
Acceptable use. You will not: (a) infringe or misappropriate IP or privacy rights; (b) use credentials, subscriptions, or institutional access in ways that violate their terms (including bulk downloading/scraping of paywalled content); (c) attempt to bypass technical controls or rate limits; (d) upload malware or interfere with the Service; or (e) process personal data you're not authorized to process.
Indemnity. You'll defend and hold us harmless from claims related to your User Content or your use of Beaver (including copyright/license claims).
Eligibility. You must be at least 13 years old (or the minimum age of digital consent in your country) to use Beaver.
3) AI models, embeddings & providers (including BYOK)
How models work. Beaver uses third‑party APIs for embeddings and AI responses. We offer two types of model options:
Beaver‑managed models. Some models offered in Beaver (such as models branded "Beaver") are managed by us. We select and route requests to one or more third‑party providers on your behalf, including automatic fallback between providers for reliability. You may not know which underlying provider handles a given request. We may route requests to any provider on the current published list. We will update the published list at least 7 days before routing requests to a newly added provider. We select providers whose API terms do not permit training on your data, and we configure available provider settings accordingly. The list of providers we currently use is published in our Privacy Policy.
User‑selected models. You may also choose a specific third‑party model/provider from the available options. Model availability may change.
Provider terms. When we send your prompts/queries and the necessary context to a provider—whether through a Beaver‑managed model or a user‑selected model—that provider's terms and data‑use rules also apply. For Beaver‑managed models, we publish the current list of eligible providers in our Privacy Policy; by using a Beaver‑managed model, you acknowledge that any provider on the current published list may process your request at any time. We will update the published list at least 7 days before routing requests to a newly added provider.
Our credits. When you use Beaver's included AI credits (whether through Beaver‑managed or user‑selected models), we configure providers not to use your data for model training where such settings are available.
Bring‑Your‑Own‑Key (BYOK). You may supply your own provider API key. We do not store your key server‑side and we do not log it; it is transmitted with each request for routing to the provider you chose. You are responsible for charges, rate limits, and data‑use rules of that provider. Providers may retain prompts or logs per their policies when you use BYOK. Review and configure your provider's data‑use and retention settings; we can't control them.
Outputs. As between you and us, you own your AI outputs to the extent permitted by law and third‑party terms. Outputs may be non‑unique and can contain errors; do not rely on them without independent verification.
4) Quality improvement, evaluation, and training
Our commitment. We never use your files and attachments ("Files") for model training, quality improvement, or evaluations. For AI Data (prompts, queries, and AI-generated outputs from both interactive and automated features), we only use this data for improvement purposes if you explicitly consent through your account settings. Operational logs are handled separately and described below.
What may be used (with your consent only). If you enable data sharing in your account settings, we may use your AI Data for purposes such as model training, feature evaluation, quality measurement, research, and the overall improvement of Beaver. AI Data includes your prompts, AI responses, and any document excerpts or page images that were retrieved during an interaction—whether from a chat conversation or an automated feature. It may also include metadata associated with library items—such as authors, titles, abstracts, and other bibliographic information. However, we do not use your original PDF files or full library attachments for training—only the specific content processed by AI features. You can change your data sharing preference at any time in your account settings.
- 30‑day window & de‑identification. AI Data used for product improvement remains linked to your account for up to 30 days to support operations, safety reviews, and user deletion requests. During that period, you may request deletion of your linked AI Data. After that, we store the data separately and remove direct identifiers (e.g., account ID/email) and use automated systems to identify and remove or replace personal information within the content itself before using it for product improvement. We do not consider references to published research, academic authors, or scholarly works to be personal information requiring removal. We cannot guarantee that all personal information will be detected and removed through this automated process.
Providers & BYOK reminder. When using Beaver-funded credits, we configure supported providers not to train on your data where such settings are available. When using your own API key (BYOK), your provider's terms and data practices apply and are outside our control. We do not store your API key; it is transmitted securely with each request.
Operational logs and history. Even if you do not share AI Data for improvement, we retain and may use:
(a) Operational logs and diagnostics — request metadata (timestamps, model ID, token count, estimated cost), error and warning details, tool call identifiers, and diagnostic context — for operational integrity, billing, dispute resolution, and regulatory compliance. These logs primarily contain request metadata. User content is stripped from these logs, though traces may remain, particularly in error messages. We use operational logs to find and fix problems. Logs are not used for product development, feature design, or model training. We do not systematically review log content for feature ideas, but fixing a bug may result in a code change that improves the service. Operational logs are temporary and are automatically and permanently deleted after 90 days.
(b) Interaction history (chat and automated feature outputs) so you can view and restore past threads. This history is visible only to you and is not used by us for product improvement. You can delete your interaction history at any time in your account settings.
5) Fair use and service limits
Fair use. We may apply caps or throttling to keep the Service stable for everyone.
Changes. Features, models, pricing, and plan limits may change. We'll give notice for material changes.
No general file hosting or backup. Beaver indexes and processes files solely to provide app features; it is not a general file‑hosting or backup service. We may throttle, restrict, or remove usage that undermines service stability or violates these Terms.
Storage management (free users of Cloud Library Sync only). For accounts with Cloud Library Sync enabled, we may periodically remove our stored copies of your synced files and derived data (such as text extracts, embeddings, and thumbnails) to keep the Service fast and affordable. This does not change or delete your original files in connected sources (such as Zotero). If you later access those items again, we may re‑fetch and re‑process them on demand, which can cause a short delay while files are restored. The Service is not a backup or general file‑hosting service—please keep your own copies.
6) Plans and Subscriptions
Plans
Beaver offers free and paid plans. Plan features, included credits, and pricing are described on our pricing page and in the app. We may modify plan offerings, features, included amounts, and pricing at any time. Material changes to paid plans will be communicated with reasonable advance notice (see Section 17).
Subscriptions
Recurring billing. Paid plans are billed on a recurring basis (e.g., monthly). By subscribing, you authorize us to charge your payment method at the start of each billing cycle at the then-current rate.
Auto-renewal. Subscriptions renew automatically until canceled. You can cancel at any time from your account settings. Cancellation takes effect at the end of the current billing period, and you retain access to paid features until then.
Price changes. We may change subscription pricing. For current subscribers, price changes take effect at the start of the next billing cycle following at least 30 days' notice. If you do not agree to a price change, you may cancel before the new billing cycle begins.
Downgrades and plan changes. If you downgrade your plan or cancel, the change takes effect at the end of your current billing period. Any included plan credits exceeding the new plan's limits may be forfeited at that time.
7) Credits and Usage
What credits are. Some Beaver features are measured using credits. Credits are a unit of service measurement that tracks how much of the Service you use, similar to how a mobile plan tracks minutes or data. For regulatory clarity: credits are not currency, stored value, or a financial product. We include this language because usage-measurement systems can otherwise be mischaracterized as stored-value products subject to money-transmitter regulations — not because we want to devalue what you've purchased. When you buy credits, you are paying for access to Service features, and we stand behind that. We may adjust credit offerings over time, subject to the notice commitments in Section 6 and applicable law.
Credit types. We may offer different credit types for different categories of usage (e.g., document processing, chat and AI operations). The available credit types and what they measure are described in the app.
How credits are consumed. The number of credits consumed depends on factors such as the type of action, features invoked, and other variables. Current credit consumption rates are always available on our pricing page. We may adjust credit consumption rates at any time; however, increases to credit consumption rates will be communicated at least 30 days in advance.
No double-charging. Work done to process documents consumes the applicable document-processing credits, not chat/AI credits. Work done to fulfill a chat or AI request consumes chat/AI credits as described in the app.
Included credits. Some plans include a periodic allocation of credits. Rollover rules, accumulation caps, and expiration terms are described on the pricing page and in the app.
Credit packs. You may purchase additional credits ("credit packs"). Purchased credits expire as stated at the time of purchase. Except as required by applicable law or as described in Section 15 (termination and service discontinuation), purchased credits are non-refundable once delivered to your account.
BYOK and credits. If you use your own API key (BYOK), you are not charged credits for the underlying model call. You may still opt-in to use additional features that require credits, as described in the app.
Debiting and failures. We do not debit credits for clear provider-side failures before any usable output is produced. If a request produces output (even partial) or if you retry a request, credits may be debited. Using the Stop button interrupts display on the client side, but processing may continue on the server or provider side — any output generated before processing actually stops may still be debited.
Pausing. If you exhaust your credits, we may pause the relevant features until credits become available (through plan renewal, purchase, or other means). Features that do not require credits, including use of your own API key, remain available.
Non-transferable; no cash value. Credits have no monetary or cash value, are not redeemable for cash, and are not transferable between accounts. Credits are forfeited upon account deletion — except as required by applicable law. See Section 15 for what happens to credits upon cancellation or account deletion.
Promotional credits. Promotional or bonus credits may have different terms, limitations, or expiration dates. They carry no cash value.
Page Credits
Grant & scope. Page Credits are a one-time allocation tied to your account and are consumed when we successfully process a page from your documents.
What counts as a "page." For PDFs and similar formats, a page is a document page. For other formats, we may use a reasonable page-equivalent.
Failures are reimbursed. If processing fails on our side (no usable output is produced), we reimburse the Page Credits for the affected pages. If a failure isn't automatically recognized, contact us within 30 days and we'll investigate and credit any confirmed errors.
No rollover/renewal. Page Credits do not renew or extend on a schedule; they persist for the life of your account and are forfeited upon account deletion.
Minimums & rounding. When a calculation results in a fraction, we round up to the nearest 0.1 Credit (and may apply a minimum debit per action as shown in the app).
8) Billing and Payments
Payment. You agree to pay all fees associated with your plan and any credit purchases. Payments are processed by our third-party payment processor. You are responsible for keeping your payment information accurate and current.
Currency. All prices are stated in US dollars unless otherwise indicated.
Taxes. Stated prices may not include applicable taxes, duties, or governmental charges. You are responsible for such charges except where we are legally required to collect and remit them on your behalf.
Refund policy. Except as required by applicable law (including mandatory consumer-protection rights) or as described in Section 15, subscription fees and purchased credits are non-refundable. Canceling a subscription does not entitle you to a prorated refund for the remainder of the current billing period.
Billing disputes. If you believe we mis-measured usage or mis-debited credits, contact us within 30 days of the charge. We'll investigate and credit your account for any confirmed errors.
EU/UK/EEA Consumer Rights
If you are a consumer in the EU, UK, or EEA:
Right of withdrawal. Under applicable consumer-protection law, you may have a 14-day right to withdraw from an online purchase. By subscribing or purchasing credits, you expressly request that the service begin and/or that credits be delivered to your account immediately, and you acknowledge that you thereby lose your statutory right of withdrawal once performance has begun or credits have been delivered to your account. Nevertheless, as a voluntary courtesy, if you contact us within 14 days of a purchase and you have not used any of the purchased credits or accessed any paid features, we will process a full refund.
Mandatory protections. Nothing in these Terms excludes or limits mandatory consumer rights under the laws of your country of residence, including rights under the EU Consumer Rights Directive, the UK Consumer Rights Act 2015, or equivalent local legislation.
9) Privacy & data handling (summary)
See our Privacy Policy for details. High level:
Where data lives. Your data handling depends on your account configuration. By default, your library and files stay on your device; we access only titles and abstracts to generate search embeddings, and your chat messages (which may include document excerpts and page images) are processed on our servers. If Cloud Library Sync is enabled, your library data and files are also synced to our servers. Server-side data is primarily processed and stored in the United States (e.g., Google Cloud Run/Functions in us-central, Supabase). See Section 1 and our Privacy Policy for details on each configuration.
Sub‑processors. We use reputable vendors such as Google Cloud, Supabase, Stripe, Voyage AI, Modal, turbopuffer, Logfire, and E2B. Sub‑processors may change; we'll update our list and provide notice of material changes when required.
Logging & monitoring. We record system logs (including user IDs, request metadata, performance metrics, and error messages) that are processed by our subprocessors (Logfire and Google Cloud Platform) and retained for up to 90 days before deletion. We also maintain operational sync logs and AI usage logs in our primary database for service integrity, billing verification, dispute resolution, and diagnosing and fixing errors (see Section 4).
Security. We apply reasonable safeguards, including encryption in transit and at rest, and access controls. However, no system is perfectly secure.
Deletion. You can request deletion of your account. Account deletion is scheduled within 10 days of your request. You'll receive an email notification when the deletion is complete. You can continue using Beaver until the deletion is processed, though you'll be signed out when your current authentication token expires. Active data is deleted within this timeframe; backups and logs are removed on a rolling schedule.
EU/UK/EEA. Controller and processor roles depend on your account configuration. For local-only users, we act as an independent controller for your account data, chat data, and the titles/abstracts we process for embeddings. For Cloud Sync users, you are typically the data controller for your synced library data and we act as your data processor; for account and operational data, we may act as an independent controller. International data transfers rely on appropriate safeguards (e.g., SCCs). A Data Processing Addendum (DPA) is available on request for Cloud Sync users.
Legitimate interests (Art. 6(1)(f)): service reliability, security/anti‑abuse, analytics, diagnosing and fixing errors, and system monitoring.
10) Non‑affiliation
Beaver is not an official service of Zotero, Microsoft, or any other third-party platform. These third parties bear no responsibility or liability for Beaver.
11) Copyright complaints
If you believe material processed through Beaver infringes your rights, email contact@beaverapp.ai with: (a) your contact information; (b) identification of the work claimed infringed; (c) identification of the infringing material; and (d) a statement, under penalty of perjury, of your good‑faith belief and your authority. We may remove or restrict access to allegedly infringing material and may terminate repeat infringers. If we remove material, we may notify the user, who may submit a counter‑notice. As we grow, we may register a formal DMCA agent if required.
12) Disclaimers
The Service is provided "as is" and "as available." We disclaim all warranties (including merchantability, fitness for a particular purpose, and non‑infringement). We don't promise uninterrupted or error‑free operation, or the accuracy of AI outputs.
13) Limitation of liability
To the fullest extent permitted by law, our total liability for anything related to Beaver is limited to the greater of US $100 or the amount you paid us in the past 3 months. We are not liable for any indirect, incidental, special, or consequential damages (such as lost profits, lost data, or business interruption). Some jurisdictions do not allow these limitations, so your statutory rights may still apply.
14) Account security
You're responsible for maintaining the confidentiality of your credentials and for activity under your account. Notify us promptly of any suspected unauthorized use or security incident.
15) Termination
You can stop using Beaver at any time. We may suspend or end access if you violate these Terms, threaten service stability/security, or as required by law. Sections that should survive (ownership, licenses, disclaimers, limits) will continue to apply.
Subscriptions and termination. If you cancel your subscription, you retain access to paid features and your included plan credits (including any rollover balance) through the end of the current billing period. At the end of that period, your subscription ends and any remaining included plan credits and rollover credits are forfeited. Purchased credit packs are not affected by subscription cancellation; they remain available until they expire or your account is deleted. Promotional credits remaining at cancellation are forfeited.
If your account is deleted (whether by you or by us), any active subscription is canceled, no refund is provided for the unused portion of the current billing period (except as required by applicable law), and all credits — included, purchased, and promotional — are forfeited. If we terminate your account for reasons other than your breach of these Terms, fraud, or abuse, we will refund the pro-rata value of any unused purchased credit packs.
Service discontinuation. If we permanently discontinue the Service, we will refund the pro-rata value of any unused purchased credit packs remaining in affected accounts at the time of discontinuation. This does not apply to included plan credits, promotional credits, or expired credits.
Tier adjustments for inactivity. If you are on a non-paying tier that includes cloud storage or processing features (such as Cloud Library Sync on a free account) and there is no service activity (such as chat messages or automated processing) for 90 days, we may migrate your account to a more limited tier and delete stored file copies and derived data (such as embeddings, processed text, and thumbnails). Before making this change, we will send notice to the email on file and give you at least 14 days to sign in and use the service to retain your current tier. Your account remains active—only the tier and associated server-side data changes. Your original files in connected sources are unaffected. This does not apply to paid subscription accounts in good standing.
Free account inactivity and deletion. We want to keep inactive, unused data from piling up. If you're on a free account and there is no service activity for 90 days, we may (i) deactivate the account or remove stored file copies and derived data, and later (ii) delete the account and remaining data. Before deleting an account, we'll send notice to the email on file and give you at least 14 days to sign in and use the service to keep it active. If your email bounces or you don't respond, we may proceed with deletion. This doesn't affect any paid subscription accounts in good standing. We may retain minimal records needed for security, fraud prevention, or legal obligations (e.g., billing records) as described in our Privacy Policy.
16) Governing law
These Terms are governed by the laws of Massachusetts, USA, excluding conflict‑of‑law rules. Mandatory consumer rights in your home country still apply.
Export controls. You must comply with U.S. and applicable international export and sanctions laws. You may not use the Service if you are subject to sanctions or in a prohibited jurisdiction.
17) Miscellaneous
These Terms are the entire agreement between you and us regarding the Service. If any provision is unenforceable, the remainder remains in effect. Our failure to enforce isn't a waiver. You may not assign these Terms without our consent; we may assign to an affiliate or in connection with a merger or sale. Neither party is liable for delays due to events beyond its reasonable control.
Changes to these Terms. We may update these Terms. Material changes will be notified via email or in‑app notice at least 15 days before they take effect (or sooner for legal/security reasons). Continuing to use the Service after the effective date means you accept the changes.
18) Contact
Questions, requests (including deletion/export): contact@beaverapp.ai