Case Studies
Anonymized snapshots of how early customers put Raina into production. Named, fully-cited case studies will be published as customers opt in to attribution.
Cutting pre-publication verification from hours to a focused review cycle
A regional investigative desk needed to corroborate mixed-media tips (user-submitted video, forwarded screenshots, and quoted social posts) before publication, without standing up a bespoke forensics lab.
Workflow
- 1.Intake: tip uploads are routed into Raina via the dashboard; editors see a single trust trace per asset.
- 2.Analysis: media forensics, C2PA provenance (when present), and source credibility run in parallel; contradictory signals are surfaced as questions, not verdicts.
- 3.Review: an editor signs off on a human-readable report that cites the evidence behind each signal before a story goes live.
Observed outcomes
- Fewer blocked stories stuck waiting on manual forensics
- Consistent methodology across desks instead of ad-hoc judgement calls
- Audit-ready report attached to every published piece
Defensible authentication of social-media and mobile evidence in eDiscovery
A litigation team needed repeatable authentication for screenshots, video clips, and forwarded messages submitted in discovery, with a record they could hand to opposing counsel and the court.
Workflow
- 1.Ingestion: case managers attach exhibits; Raina hashes, timestamps, and logs chain-of-custody at every step.
- 2.Analysis: each exhibit is scored for manipulation likelihood and provenance; conflicting signals are preserved rather than resolved away.
- 3.Export: a court-ready bundle with methodology notes, signal breakdown, and verification URL is generated for deposition and filing.
Observed outcomes
- Single verification record per exhibit, linkable by certificate ID
- Clear separation between what the system observed and what it inferred
- Reduced expert-witness preparation time for routine authenticity questions
Catching synthetic documents and voice-cloned instructions before wire release
A fraud-operations team was seeing an increase in AI-generated identity documents and voice messages impersonating clients authorizing transfers. They needed a secondary check that sat outside the primary KYC pipeline.
Workflow
- 1.Trigger: high-risk wires and onboarding packets are mirrored to Raina via API.
- 2.Analysis: document forensics flag regeneration artifacts and template reuse; voice clips are checked for known synthesis signatures.
- 3.Escalation: suspicious packets are routed to a human reviewer with the specific signals that flagged, not just a score.
Observed outcomes
- Explainable review queue: analysts see why something flagged
- Lower false-positive load compared with score-only tools
- Clean audit trail for regulator inquiries
Rapid triage of viral content claiming to originate from the agency
A communications office needed a fast, defensible way to triage videos and screenshots circulating online that appeared to come from agency officials, without making public statements based on gut calls.
Workflow
- 1.Monitoring: flagged URLs and uploads are sent to Raina as they are reported by staff or the public.
- 2.Analysis: provenance, manipulation indicators, and source credibility run in minutes; the trust trace and evidence bundle are preserved.
- 3.Response: the office decides whether to issue a correction, with the verification record attached to the internal decision log.
Observed outcomes
- Shortened path from "is this real?" to a defensible decision
- Internal-facing record of what was checked and when
- Consistent response framework across incidents
Open to being featured by name?
Contact our team