Effective September 20, 2026. This policy explains what we collect, how we handle it, and who touches it. The short version: your case files exist here to be transcribed, summarized, or converted, and for no other reason.
Account information (name or firm, email, a hashed password). The case materials you upload, plus the case details you enter on an order (caption, county, add-on choices, notes). Order and payment records. Basic site analytics. We never see or store your card number; payment details go directly to Stripe.
Files are uploaded over an encrypted connection and stored encrypted at rest on our own equipment in Abilene, Texas. Transcription runs on our in-house hardware, not on a public-cloud transcription service. Files are hashed at intake so every deliverable traces to the exact source you sent. Access is limited to the owner, the systems that process your order, and the staff or systems assigned to produce the deliverable.
Speech recognition and document conversion run locally on our machines. Correction, quality review, and discovery summaries additionally use Anthropic's Claude models over an encrypted connection, under commercial terms that do not permit training on your content. We never use your files to train models, never sell your data, and never share it with anyone who is not part of producing your deliverable.
Stripe processes payments. Cloudflare carries our web traffic and provides the human check on signup. Anthropic provides the AI models described above. Google provides site analytics and fonts, and Gmail carries our transactional email (order notices never include your case files). Each provider receives only what its role requires.
If you connect your account to Mehaffey Discovery in Codex or ChatGPT, you choose permissions on this website. We keep connection metadata, hashed access credentials, granted scopes and access audit records. Your website password is not shared with your AI client. Depending on your permissions and requests, Codex or ChatGPT can receive your display name, order and billing status, case details, file names, source identifiers, discovery text, transcripts, summaries and supported deliverables. OpenAI processes information received by Codex or ChatGPT under the terms and settings of your OpenAI account. Check applicable protective orders and confidentiality requirements before authorizing access. You can revoke a connection from Connected apps; revocation prevents future access but does not erase information already received by your AI client.
The discovery hub reads existing order records without copying evidence. An explicit import into Form Builder creates separate workspace discovery snapshots; these are not automatically deleted when a transcription order reaches its retention deadline. Manage those snapshots in Form Builder or contact us about deletion. Connection revocation does not delete them.
Source files and deliverables are automatically deleted 30 days after delivery. We email a reminder a week before, and a certificate of destruction automatically after, listing exactly what was destroyed. Account and order records (dates, amounts, file names, hashes) are kept for accounting. To delete your account and its records, email us; we will keep only what accounting and law require.
If your discovery is subject to a protective order, tell us on the order form and send the order; we handle the materials accordingly and sign acknowledgments where required. We treat all client material as confidential whether or not an order requires it.
A session cookie keeps you signed in. Google Analytics sets its own cookies to show us aggregate site usage. No advertising cookies, no cross-site tracking of ours.
You can download your deliverables any time before the retention window ends, request earlier deletion of an order's files, request a destruction certificate, or ask what we hold about your account. Write to [email protected] and we will respond promptly.
Our services are for legal professionals and their clients, not children. If we change this policy, the effective date above changes with it. Questions or requests: [email protected], or Mehaffey Transcription, 2441 South 1st Street, Abilene, TX 79605.