Privacy Policy

LaunchCut at opportunityintel.top is operated by matt, who is responsible for the personal information processed for the Service. Contact us for privacy questions, access requests or account deletion at matt1232026@gmail.com.

This Policy covers the website, local editing tools, account workspace and communications with us. Third parties have their own notices for services they provide directly to you. This Policy describes current processing; it does not grant permission to introduce unrelated uses of your information.

In Creative Studio and Impact, your selected recording is decoded, sampled, cropped, composed and encoded in your browser. The recording, extracted frames and locally generated export are not uploaded to LaunchCut simply because you select or edit a file. Local processing may access images, voices or text already in that file; we do not receive those contents through this workflow.

Browser storage can hold your language preference, last workspace reference, story text, scene timings and editing settings. Creative Studio drafts also hold a recording’s file name, size, duration and a fingerprint calculated from small portions of the file to recognize it when reselected. These draft details remain on your device unless you separately choose to send them to the account workspace or support. Browser storage does not retain the recording itself.

You can remove local drafts and preferences by clearing this site’s browser storage. Closing a page releases its working media buffers; downloaded exports remain wherever you saved them. Your browser, operating system, extensions and any device backup service are outside our control. Keep your own backups and avoid editing sensitive material on a shared device.

Website requests

Cloudflare receives the IP address, requested URL, request time, browser/network headers and necessary security information when it serves the website or API. We use operational logs to troubleshoot errors and protect the Service; we do not use them for behavioral advertising.

Email sign-in

We receive the address you enter and the language used for the verification email. The backend stores an account identity, display name, login state and hashed verification/session data. Email codes are time-limited; we do not ask you for an email-provider password.

GitHub sign-in

If you choose GitHub, we request read:user and use your GitHub numeric account ID and username to create or recognize your account. An access token is used during the identity exchange and is not saved as a reusable token in our account database. We do not request repository access or store your GitHub password.

Account workspace

We receive project names, briefs, scripts, scene settings and other information you expressly save to the backend, along with project IDs, timestamps, account credit balances and job or asset metadata where relevant. A local studio draft is separate from a saved backend project.

Support

We receive your contact details, message and attachments you choose to email us. Please redact sensitive content and do not send sign-in codes, API keys or complete payment credentials.

Payment records, if enabled

The payment provider collects the checkout details. Our backend records order, checkout and payment identifiers, the selected pack, credit quantity, payment status and transaction time. We do not store complete card numbers, card security codes or banking passwords.

Local export billing

Even though your recording stays on your device, the backend receives your product name, studio identifier and export request, reservation and result status. It stores an export ID, timestamps, credit debits or returns and order references to maintain your balance and activity history. No recording frames, local file content or exported MP4 are sent with these billing requests.

One-time credit purchases are available through Waffo. Google sign-in and full recording cloud uploads are not currently enabled. Optional OpenAI generation is described below. A payment-test flow may create test transaction records. Sign-in is required for local editing; exports and downloads are included; AI creation requires 30 available credits. Public registrations do not receive free credits.

Before AI creation, your browser extracts up to 16 compressed keyframes locally. You can inspect them and uncheck private frames. Only after you check the consent box and confirm the displayed charge (30 credits for Standard or 60 for Enhanced) are the selected JPEG frames, product name, optional brief, language, recording dimensions, duration and a file fingerprint sent through the LaunchCut backend to OpenAI. The full recording, audio track and finished MP4 are not sent. One selected frame may also be used as visual context for generating one campaign image.

We do not save the submitted keyframe images in our database or R2. They pass through working memory to deliver the task. We retain the generated storyboard and image in your protected backend account for up to 24 hours to recover the same generation, then make them unavailable and remove their content during cleanup. Task identifiers, product name, status, timestamps and technical usage metadata may be kept for up to 90 days for support and limits; credit ledger records follow the transaction retention described below. The browser saves the generated image locally in IndexedDB.

OpenAI is an external provider and may process this content outside your country, including in the United States. We request Responses with store=false. This does not provide zero retention: OpenAI normally retains abuse-monitoring content for up to 30 days and may retain certain content longer for legal or safety reasons. OpenAI states that API data is not used for model training by default unless the API account opts in; we do not opt in through this feature. See OpenAI API data controls ↗. Contact us at matt1232026@gmail.com for privacy or deletion requests.

You may decline AI processing and keep using manual editing and preview. Do not send credentials, private messages, financial data, health records or other sensitive content. Consent applies only to the confirmed task; you can stop future submissions. Withdrawing consent cannot undo processing already completed. Review generated text and images before publication.

The same selected keyframes and product description also undergo OpenAI content checks before generation. Generated storyboard text and campaign imagery undergo checks before a result is released. We retain only classifications, policy version, task ID and check timestamps for up to 90 days, without copying prompt text or keyframe images into those check records. Review reports and appeals are handled by the operator through the support email. No additional provider receives your recording for these checks.

Enhanced video: after you separately confirm Enhanced, the planned shot prompt and generated campaign reference image are sent through our backend to Pixstag to create one five-second clip. The full recording and original UI frames are not sent to Pixstag. Generated poster and last-frame images are sent to OpenAI for content, composition and editing checks. We keep the provider task ID and generation metadata to resume the same task and deliver the result, with access restricted to your account. The generated clip passes through our backend to your browser and can be cached locally for the work’s 24-hour period. Our backend does not retain a copy of the full generated video. Pixstag processes and stores inputs and outputs under its own published terms; their expiry and deletion are independent of LaunchCut’s 24-hour work expiry, which does not delete provider copies. Its service and contact details are at Pixstag ↗. Contact us before using Enhanced if its processing is not acceptable to you.

We use account identifiers to authenticate you and maintain your workspace; email addresses and verification messages to deliver sign-in codes; project information to save and return the projects you request; export billing records to authorize and settle exports; and order records to reconcile payments and entitlements if purchases are enabled. Support messages help us answer your request.

Security cookies, limited login-attempt counters and operational information help prevent unauthorized access, code abuse and fraud, and diagnose faults. We may retain or disclose specific records when required by law or to resolve a substantiated dispute. We do not use your recordings for advertising profiles, face recognition, biometric identification or training our own AI models.

Where applicable law requires a legal basis, core account and workspace processing is necessary to provide the service you request; legally required records are processed for compliance; proportionate security and troubleshooting may rely on legitimate interests where that basis is recognized and your rights are considered. Optional processing relies on consent when required. Reading this Policy alone is not consent, and nonessential processing must not be made a condition of unrelated local editing.

We keep daily aggregate counts of public-page visits, example plays, creation-entry clicks, recording selections, completed previews and exports. The category of the referring source may be kept in tab session storage until that tab session ends. Each step is counted at most once per page and day in that tab session. These approximate counters do not identify unique people or prove a sale.

The measurement endpoint accepts only predefined event, page and source categories. It does not store your account, email, IP address, recording, prompts, full URL or referrer as analytics data. Daily measurement rows are retained for up to 30 days. Separate security rate-limit counters protect the endpoint. Browser Do Not Track and Global Privacy Control signals disable this optional measurement. These counters are not used for advertising profiles. Payment and generation totals are separately calculated from service records.

We use essential authentication cookies and browser storage for features you choose. We do not currently install advertising pixels, cross-site marketing cookies or a third-party audience analytics script.

__Host-launchcut

Secure, HttpOnly session cookie used to keep you signed in. Its maximum lifetime is 30 days. Signing out invalidates the active session.

__Host-launchcut-auth

Secure, HttpOnly login-browser cookie used with a request token to protect email authentication. Its maximum lifetime is 30 minutes.

launchcut_oauth_github

Temporary cookie used to validate a GitHub authorization response. Maximum lifetime: 10 minutes. It is cleared during callback. Google’s equivalent is only relevant if Google sign-in is later activated.

Local browser storage

Generated campaign images and optional product logos remain in IndexedDB until you clear site data. The selected logo is used locally for the opening and/or ending and is not sent to OpenAI by the logo feature. Removing or replacing it deletes the current stored logo copy. Downloaded storyboards may include the logo; protect those files as your own media. Language preferences, editing drafts and a last-project reference remain until overwritten or cleared. A temporary return-to-studio path is saved in session storage for up to 30 minutes during sign-in. They are used to restore your chosen settings, not for advertising.

You can block cookies or clear storage in your browser. Blocking authentication cookies prevents sign-in; removing local storage deletes local drafts and preferences. Public example films and pricing can be viewed without an account. Editing studios and credit activity require sign-in. We do not change our no-ad-tracking practice based on a Do Not Track or Global Privacy Control signal. If nonessential analytics or advertising is introduced, we will update this notice and provide the choices required by law before enabling it.

Only information needed for the relevant function is shared with these providers. The linked notices include their identity, privacy contacts, international processing and procedures for requests concerning data they control.

Cloudflare

Website delivery, backend hosting, security and account/project database storage. It processes request information and backend records. Cloudflare is a US-based global provider; processing is not limited to a server in your country. Cloudflare privacy notice ↗

Resend

Delivery of sign-in emails. We send your address, message language and verification-message content, including the temporary code. Stored email data is held in the United States under Resend’s service policies. Resend privacy notice ↗ · Retention and security information ↗

GitHub

Optional OAuth authentication. GitHub receives authorization and browser information and returns the limited profile data described above. Its processing may occur in the United States and other locations. You can revoke the LaunchCut app in your GitHub account settings. GitHub privacy notice ↗

Google Fonts

Delivery of the fonts used on the website. When your browser requests remote font styles and files, Google receives request information such as your IP address and browser headers. This does not enable Google account sign-in. Google privacy notice ↗

OpenAI and Pixstag — optional AI creation

OpenAI receives only the selected keyframes and product information you confirm. Enhanced sends the planned motion prompt and generated reference image to Pixstag; full recordings are not sent. The generated video preview frames are reviewed by OpenAI. See the AI data processing and provider retention section for details.

Waffo Pancake — credit purchases

When you choose Waffo checkout, we send your account identifier, selected product, price, order reference and checkout language to link the purchase to your account. Waffo collects checkout billing information and returns signed payment or refund records. Production checkout processes real one-time purchases; clearly marked isolated test checkouts do not collect real money. Waffo privacy policy ↗ · Waffo buyer terms ↗

Dodo Payments — test / future checkout

Checkout and payment administration only when you open that checkout. Dodo may independently process billing, transaction, fraud-prevention and tax information as merchant of record. Dodo is available only when explicitly enabled for live checkout. Dodo privacy notice ↗ · Roles and data processing ↗

If you email our Gmail contact, Google processes that message for email delivery and mailbox storage under its own terms. Full recordings stay on your device. If you confirm optional AI creation, selected frames and text are sent to OpenAI as described below. Social platforms receive an export only if you publish it there yourself.

We do not sell or rent personal information, share it for cross-context behavioral advertising, or make your local recordings public. We do not publish your submitted project or support attachment as a case study without separate permission.

We may disclose a limited amount of relevant information when a valid legal obligation requires it, to investigate a substantiated security incident or defend a legal claim, subject to applicable law. A future transfer of the Service would require notice of the recipient and continued protection of your information; any materially different purpose would need an appropriate legal basis and consent where required.

LaunchCut’s operator is based in mainland China, and the hosting, email and authentication providers above operate internationally. Account identifiers, sign-in messages, support emails and network requests may be processed outside your country, including the United States. This statement does not mean your local recording is uploaded or transferred to those providers.

Different jurisdictions may have different privacy protections. Where applicable law requires transfer notices, separate consent, an assessment, a contractual mechanism or another safeguard, those requirements must be addressed for the relevant processing. Acceptance of the Terms is not a substitute for separate consent. This Policy does not claim that a regulatory certification or a particular transfer filing has been completed.

You can contact us to ask which recipients handle your account information, request available information about safeguards, exercise your rights, or choose to use the local tools without creating an account. Provider privacy links in the previous section explain how to contact them directly about information they control.

Local recordings and exports

Working media remains on your device. Generated works, artwork and browser-saved MP4 expire 24 hours after creation and are cleared while the website is open or on its next visit. Downloaded exports follow your own file deletion and backup settings. We cannot delete a file from your device remotely.

Verification codes and OAuth state

Email codes and OAuth state expire after 10 minutes. Expired verification and temporary authentication records are removed by scheduled cleanup. A code digest may exist until that cleanup; no raw email code is stored in our verification table.

Sessions and abuse limits

Sessions expire after at most 30 days; temporary browser records expire after 30 minutes. Login rate-limit records use hashed keys and short counting windows of up to a day. Expired records are removed during cleanup.

Account and saved projects

Kept while the account and projects are in use, until you delete the project or request account closure. We aim to remove associated active account/project records within 30 days after a verified deletion request, unless a specific legal or security reason requires retention.

Support and operational records

Support correspondence is retained while resolving your request and for up to 12 months after closure unless needed for an ongoing dispute or legal obligation. Operational logs follow the hosting provider’s limited log-retention settings; records extracted for a security incident are kept only as long as needed to investigate or meet an applicable obligation.

Transactions, if purchases are enabled

Necessary payment and credit-ledger records may outlast account closure for accounting, tax, dispute and fraud obligations. The applicable period depends on the record and governing law; we will identify the applicable basis and period on request. The selected payment provider retains its own records under its notice.

Resend’s current Free-plan service documentation states that email and log data is retained for 30 days. Deletion from our active database does not instantly remove a provider’s lawfully retained records or every recovery copy. Such remaining data must be restricted to its necessary purpose and removed under the relevant retention cycle.

Cloud media uploads are not offered by the current local studios. If enabled later, the existing backend expiry configuration is 7 days for source media, 30 days for rendered output and 1 day for voice previews; the upload interface and this Policy must state the actual period before use. Do not treat this configuration as a currently available storage promise.

We use HTTPS, access checks for account projects, HttpOnly/Secure session cookies, hashed session identifiers and verification-code digests, temporary codes, request protections and login rate limits. Access to backend information is limited to operating, supporting and securing the Service. A local edit does not require uploading your media.

No online service or device can be guaranteed completely secure. Secure your own device and account, do not share sign-in codes, and mask confidential information in recordings. If we learn of a personal-information incident requiring notification, we will assess it, take corrective action and notify affected people and authorities as required by applicable law.

Depending on the law that applies, you may request access to and a copy of your information, correction, deletion, account closure, restriction or objection to processing, transfer of information where applicable, or an explanation of the processing. Where we rely on consent, you may withdraw it without changing the lawfulness of earlier processing. We do not make solely automated decisions about you with legal or similarly significant effects in this preview.

Email matt1232026@gmail.com with the request type and your account identifier. You do not need to send a government identity document by default; we will ask only for proportionate verification needed to protect the account. We aim to respond within 30 days, or sooner if law requires. If a lawful extension, retained record or refusal applies, we will explain the reason and available next steps. Requests are normally free; any legally permitted fee or refusal for manifestly unfounded or excessive requests will be explained first.

Local drafts can be removed by clearing site data in your browser. You may revoke GitHub authorization separately. Closing the account does not delete downloads or posts on another platform. You may also complain to the competent data-protection authority or seek a remedy available under applicable law. We do not discriminate against you for exercising your rights.

The Service is intended for adults aged 18 or older and is not designed for children. If you believe a child has provided account information without appropriate authorization, contact matt1232026@gmail.com so we can investigate and remove information where required.

Do not send us unnecessary sensitive information, such as government IDs, precise location, health records, financial credentials or intimate recordings. For any person shown in material you edit or submit, ensure that you have permission or another lawful basis and protect their privacy. Displaying a face locally does not mean LaunchCut performs biometric identification.

The effective date and version at the top identify this notice. We will highlight material changes in the Service or use available account contact details before a new processing purpose or material change takes effect, except where urgent legal or security circumstances require otherwise. If new consent is required, we will request it before the relevant optional processing begins.

English and Chinese versions have the same intended meaning. You may ask us to clarify any wording or processing practice.

Operator: matt
Privacy and support email: matt1232026@gmail.com