Frequently asked questions.

Any civil litigation matter involving document discovery. Commercial disputes, employment litigation, insurance defense, IP disputes, personal injury with large document volumes. If your firm processes discovery documents, Auris fits your workflow.

Three-layer case intake: (1) structured case profile with party names, case type, date range, key issues; (2) complaint/petition upload where the AI extracts legal theories and factual allegations; (3) post-first-pass refinement where attorney overrides on a sample set retrain the classification boundary. The more context you provide, the more precise the classifications.

In testing across 600+ documents including real Enron corpus data and adversarial edge cases: 98% relevance accuracy, 100% privilege accuracy, zero false negatives on relevance. Zero false negatives is our non-negotiable standard - if a document could be relevant, Auris flags it.

No. Auris uses paralegal observation language only. Documents are classified as “Likely Relevant” or “Privilege Indicators Detected” - never “Relevant” or “Privileged.” The attorney always makes the final call. Every AI classification is surfaced with its reasoning for human review.

$5,000 one-time implementation, $3,000/month subscription, $5,000 annual maintenance. Year 1 total: $46,000. For comparison, a 50,000-document case reviewed manually at $80/hour paralegal rates costs approximately $40,000 in review time alone. Auris processes that same volume in hours.

All data is stored on US-based infrastructure with per-client isolation. Documents are stored in encrypted, access-controlled buckets. AI processing uses Anthropic's Claude API with zero data retention - your documents are never used to train AI models.

Yes. Auris exports Excel review summaries, Bates-numbered PDF production sets, privilege logs, and CSV formats compatible with Relativity load file standards. Auris fits upstream in your workflow - first-pass classification before documents enter your existing review platform.

Auris is in beta with a production launch targeted for May 2026. We're accepting a limited number of pilot clients now to configure the platform for real litigation workflows. Pilot participants get direct founder access and preferential pricing.

Relativity is a $50K+/year enterprise platform designed for large law firms and corporations. Logikcull is an eDiscovery tool, not an AI classifier. Auris is purpose-built specifically for first-pass relevance and privilege classification for small-to-midsize litigation firms (2-20 attorneys) at a fraction of the cost. No seat fees, no per-GB charges.

General AI tools lack document ingestion pipelines, privilege detection logic, confidence scoring, audit trails, per-client data isolation, and legal-format exports. Auris is purpose-built for the specific workflow of first-pass discovery review - from upload through classification to attorney sign-off and production export.