Slate
SECURITY & TRUST

How Slate handles your cut, your bids, and everything in between.

This page is maintained by the Slate team to answer the security and privacy questions we get most often. It reflects how the app runs today — not a certification. If something here matters for your studio and it's not covered, email us and we'll answer directly.

Last reviewed: 20 October 2026 · Implementation summary, not an independent audit or certification.

Hosting & encryption

  • Slate uses managed cloud application hosting, database, and object storage. Confirm current deployment regions and subprocessors with us before accepting residency requirements.
  • The hosted Slate application is served over HTTPS.
  • Infrastructure encryption at rest is provided by the hosting platform; script content also has application-layer encryption as described below.
  • Backup frequency, retention, and available restore options must be confirmed for the deployment as part of studio onboarding.

Access control & invites

  • Slate is invite-only during beta. Anyone can request access at /request-access; only the Slate team can approve a request and issue an invite email.
  • Users belong to workspaces, with project access resolved separately. Workspace ownership does not by itself describe every sharing or visibility permission.
  • Workspace and project access checks, together with database row-level security, are designed to isolate tenant records. These controls are not a guarantee that vulnerabilities cannot occur.
  • Workspace and project permissions distinguish editing from viewing. Authorized administrators manage membership and project access.

Bid Desk & financial records

  • Bid Desk access is granted per workspace by platform administrators. Server-side entitlement and workspace/project authorization checks apply independently of the project's vendor/client perspective.
  • Sent bid requests retain their reviewed shot scope. Scope changes require a revision rather than silently changing what a vendor received.
  • Vendor returns create quote revisions, not awards. Awards, change-order approval, and purchase-order actions remain explicit internal decisions; producer alerts do not approve a bid.
  • Captured financial baselines and historical purchase-order/change-order records are retained. New quotes do not rewrite previously approved prices.
  • Original vendor returns and generated PO documents use private, versioned artifacts. Download paths authorize access before issuing a time-limited link; support downloads have access-event records.

Vendor portals & recipient links

  • An invitation link is scoped to its recipient request and reviewed bid-pack scope, not to unrestricted workspace access or every request for that vendor.
  • Portal responses use defined, allowlisted fields rather than exposing internal workspace records. PO downloads are checked against the request's package and vendor.
  • Portal links are bearer credentials: someone receiving a forwarded link may act as that recipient while it remains valid. Keep links private and contact the producer if a link is exposed; this is not identity-verified vendor sign-in.
  • Internal portal previews use authorized workspace access without using the vendor token or recording a vendor view. Workbook preview and final submission are separate steps.
  • Submitted workbooks retain the original file and create a new quote revision. Vendor-added items require internal review before entering canonical scope; submissions do not approve awards or change orders.

Media handling

  • Production QuickTimes and captured frames use private storage with access-controlled, time-limited signed links. Treat a signed link as a credential: anyone holding it may use it until it expires.
  • Reference-video and shot-thumbnail storage uses uploader-scoped access policies. Intentionally shared onboarding/demo assets have separate policies; they must not contain confidential production material.
  • Storage access and application authorization are separate checks. Slate is intended for proxies, thumbnails, and light deliverables, not studio master-media hosting.
  • Local-only QuickTime use avoids upload. Explicit AI actions, such as cropped-frame fallback, may still send selected content to a provider.
  • Deletion and retention requirements, including backup copies and shared references, should be agreed before uploading sensitive production material.

Script handling

  • Scripts stored in a workspace are encrypted at rest with AES-256-GCM. The encryption key is kept outside the database, so a database breach alone would expose only scrambled text.
  • The script storage path encrypts page text, imported notes/highlights, and original PDFs using per-document keys wrapped with a server-held master key. Generated breakdowns, titles, metadata, and exports are separate records; do not assume every derived field has this extra encryption.
  • Keeping a script on your device avoids workspace storage, but explicitly invoked AI or import actions can transmit content. Do not invoke them for unreleased material without approval.

Dedicated infrastructure (paid upgrade)

  • The default deployment is multi-tenant, using application authorization and database row-level security. Whether it satisfies your studio requirements must be assessed against your policies.
  • Dedicated infrastructure, residency requirements, and contractual controls can be discussed as a separately scoped engagement; they are not enabled in the standard deployment.
  • Available hosting regions, customer-cloud feasibility, backups, monitoring, and operating responsibilities require technical review and written agreement.
  • This is a paid upgrade sold alongside the customization work below — talk to us about your studio's compliance and residency requirements and we'll scope it.

AI data handling

  • AI features only send data to a model provider when you explicitly trigger the action. Slate does not background-scan your projects.
  • What's sent is scoped to that action: the marker's captured frame, the shot code, and any note or transcript text you asked to be considered.
  • Provider training, retention, processing region, and subprocessor terms vary by provider, model, account, and configuration. Verify applicable terms before sending confidential material; this page does not promise zero retention.
  • Optional BYO key (available now in workspace settings): paste your own OpenAI, Anthropic, or Google key at the org level and requests bill to your provider account instead of ours. The raw key is encrypted at rest, only decrypted server-side at call time, and never returned to the browser.
  • Frame text recognition for auto-markers runs locally in your browser first; a model provider is only called as a fallback for frames the local pass can't read confidently.
  • AI usage records include action, workspace, and usage information so administrators can review spend. AI usage logs are not a substitute for reviewing provider retention; generated results may be saved as project records.

AI levels & your NDA

  • Gated AI features use workspace levels, and projects can tighten them: Off, Metadata only, Metadata + media, or Full. Those features enforce the level server-side. This is not a universal outbound-data switch; the separate Bid Desk PDF-import limitation is described below.
  • Metadata features may send your question and relevant project names, notes, shot identifiers, and numeric values. Metadata may itself be confidential under an NDA.
  • Metadata + media adds burn-in reading and dictation: a cropped burn-in strip or the audio clip you just recorded, sent only when you press the button. Playing, scrubbing, marking and bidding a QuickTime never calls a model at any level.
  • Full permits gated script features to send screenplay content for breakdown and audit. Bid Desk AI PDF-return parsing is a separate document-processing path; do not assume the script AI-level setting blocks it. Avoid that action unless your studio has approved the provider.
  • What an NDA usually turns on: whether third-party processors need prior written approval, whether processing is restricted to a region, whether metadata about the material is treated differently from the material itself, and whether every downstream vendor needs its own signed NDA or DPA. Business-tier model terms cover training and retention — they are not a signed NDA with your client.
  • If approval is unclear, use AI off and avoid AI document-import actions. Enable only the approved material classes after confirming your NDA, provider terms, and project configuration.

Private & dedicated AI

  • BYO key: supported AI paths can use workspace OpenAI, Anthropic, or Google credentials. A BYO key does not make processing local, guarantee residency, or automatically establish a DPA. Confirm routing and provider terms for each feature.
  • Dedicated inference is a possible separately scoped engagement, not an active isolation mode of the standard product. Endpoint availability, models, region, and routing require validation.
  • Customer-cloud or air-gapped operation requires a separate feasibility assessment and written scope; no air-gapped assurance is made for the standard deployment.
  • QuickTime and scripts are treated differently: playing, scrubbing or bidding a QuickTime never calls a model. Script breakdown and vendor-return parsing are the features that send documents, so those are the ones to evaluate against your NDA and choose an isolation tier for.

A BYO key changes the provider account, not the confidentiality obligations. Confirm each feature’s routing and written approvals before using confidential material. Dedicated arrangements require a separate agreement.

Audit trail & recovery

  • Edits to bids, shots, markers, and tags are recorded in a per-project activity log with the acting user, the field changed, and the before/after values.
  • Users with editor access or higher can undo their team's recent changes from that log for up to 7 days; older entries stay readable as history.
  • Locked bid versions are enforced in the database, not just the UI — a locked version can't be edited even by a crafted request.

What we don't do

  • We don't sell, share, or license your projects, cuts, or bids to anyone.
  • Workspace access requires an authenticated account and applicable access approval; vendor recipient links use a separate, limited token-access model.
  • Product analytics and operational logs should be included in your privacy review; this page is not a cookie or tracking audit.
  • We don't publish a compliance badge. If you need SOC 2, HIPAA, GDPR-DPA, or a signed security questionnaire, email us — we'll tell you honestly where we are.

Reporting a vulnerability

  • If you think you've found a security issue in Slate, please email info@whatifi.io before disclosing publicly.
  • We aim to acknowledge reports within two business days and will keep you posted while we investigate.
  • We don't run a paid bounty yet, but we'll credit reporters who ask for it.
Still have a question?

Email info@whatifi.io and a human will get back to you — usually the same day.