Native macOS Application

Stop paying $3,000/month to redact footage
you can't legally upload anyway.

Blurr batch-processes bodycam, interview, and dashcam footage on your Mac in minutes — faces, screens, license plates, and spoken names redacted automatically. No cloud. No compliance risk. No subscription treadmill.

  • Processes a 1-hour clip in under 15 minutes using Apple Silicon — walk away, come back to done
  • Originals never touched. Full JSON audit log generated automatically for every batch
200+Security teams
50k+Videos redacted
$0Cloud fees paid
Trusted by law firms, school districts, and security consultancies — limited early access pricing now open
based on early access users
⚠️AI-assisted redaction — always review outputs before sharing. Automated detection is not 100% accurate. Blurr flags low-confidence frames for manual review. Final responsibility for disclosure compliance rests with your team.

Why Law Enforcement & Healthcare Choose Blurr

Uploading raw bodycam or patient footage to a cloud redaction service introduces massive compliance risks and data sovereignty issues.

  • Zero Cloud: Your files never leave your Mac.
  • Fixed Cost: Pay once. Stop paying $0.50/minute to cloud providers.
  • Apple Silicon: Leverages the Neural Engine for blazing fast local inference.
  • Non-Destructive: Originals are never altered. Redacted copies are saved automatically.
Cam_042_Main.mp4Redacted
Interview_09.movRedacted
Dashcam_Front.mp4Processing Face & Audio...

Up and running in 3 steps

No configuration wizards. No API keys. No training required.

1
📂

Drag your footage in

Drop a folder or individual clips onto Blurr. Supports MP4, MOV, MKV, and more. Mixed batches work fine.

2
⚙️

Configure once

Choose what to redact — faces, screens, license plates, spoken names. Pick blur strength. Settings persist across sessions.

3

Walk away, come back to done

Hit Process All. Blurr handles your entire batch unattended using Apple Silicon. Redacted files appear in a /Redacted subfolder with a full audit log.

Everything you need. Nothing you don't.

From complex environments to moving vehicles, Blurr has a specialized model for the job.

👤 Face Redaction

Detect and blur faces across thousands of frames in seconds using Apple Vision.

Learn more →

💻 Screen & Text

Automatically detect monitors, laptops, and readable ambient text in the environment.

Learn more →

🔊 Audio Redaction

Remove identifying PII from audio tracks using local Whisper transcription and NLP.

Learn more →

🚗 License Plates

Identify and redact vehicle license plates to maintain anonymity in public traffic recordings.

Learn more →

Built for the teams who need it most

Blurr is trusted across industries where data sovereignty isn't optional.

🚔
Law Enforcement

Bodycam & Dashcam

Departments processing FOIA requests and court submissions redact footage locally — no third-party cloud ever touches evidence footage. Chain of custody stays intact, and the JSON audit log satisfies evidentiary requirements out of the box.

🏥
Healthcare

Patient & Training Footage

Hospitals and training programs redact patient identifiers from clinical recordings before sharing with staff or residents. HIPAA-relevant data never crosses a network boundary — because there is no network call to make.

⚖️
Legal

Deposition & Interview Video

Law firms and PI companies redact witness faces and on-screen case materials from deposition recordings before disclosure. Fixed per-seat pricing replaces unpredictable per-minute cloud billing that scales with caseload.

🛡️
Insurance

Claims & Investigation Footage

Adjusters and SIU teams redact claimant faces from surveillance and dashcam footage before sharing with external counsel or regulators. Blurr's batch queue handles high-volume claims periods without per-clip cost spikes.

Built for regulated environments

Blurr's local-only architecture is purpose-built for the compliance frameworks your legal and IT teams care about.

HIPAA

Healthcare footage stays on-premise

No PHI ever crosses a network boundary. Blurr's zero-upload architecture means patient video data never touches a covered entity's network perimeter — satisfying HIPAA technical safeguard requirements without additional BAAs or cloud DPA agreements.

CJIS

Criminal justice data stays local

CJIS Security Policy requires that criminal justice information be processed only in authorized environments. Blurr processes everything on your department-managed Mac — no third-party data processor ever touches evidence footage.

FOIA

Audit trail for every release

Every batch produces a structured JSON audit log — source filename, detection type, confidence score, model version, and timestamp. Satisfies evidentiary and disclosure requirements for FOIA responses out of the box.

Blurr is a tool, not a compliance program. Consult your legal counsel before relying solely on any software for regulatory compliance. These descriptions reflect architectural properties, not legal certifications.

Security & Privacy — the full picture

For IT and legal teams evaluating procurement. Forward this section.

🔑

No API keys required

Detection runs entirely via Apple Vision and Core ML on-device. There are no external model endpoints, no credentials to manage, and no third-party inference calls.

🌐

No network calls during processing

Blurr makes zero outbound network calls while processing footage. You can confirm this with any network monitor. The app works fully offline after installation.

📵

No telemetry on your footage

Crash reports (opt-in) contain stack traces only — never file names, paths, frame content, or any metadata from your video files. Your footage is invisible to us by design.

📋

Audit log format publicly documented

Every batch produces a structured JSON audit log with source filename, detection counts, confidence thresholds, model version, and timestamps. Schema is documented and machine-readable for compliance workflows.

🔒 Architecture principle: Nothing leaves the device. The Neural Engine on Apple Silicon handles detection at a fraction of the cost of cloud services — with zero data exposure.

Simple pricing. Devastating math.

A security firm processing 500 hours/month via cloud redaction at $0.10/min pays ~$3,000/month. Blurr Team is $79/year.

Monthly
Yearly2 MONTHS FREE
Free Trial
$0
forever
  • Fully functional
  • Videos under 60 seconds
  • No account required
Download Free
Most Popular
Solo
$29/mo
Cancel anytime · 1 seat
  • Unlimited video length
  • All redaction modes
  • Full audit log
  • Single machine
Get Solo License
Team
$79/mo
Cancel anytime · 5 seats
  • Everything in Solo
  • 5 machine licenses
  • Priority support
  • Volume processing
Get Team License
The math: Cloud redaction at $0.10/min × 500 hrs/month = $3,000–$15,000/month — before addressing the fact that many customers legally cannot upload raw footage to any third-party cloud. Blurr Team: $79/mo.

Need more than 10 seats?

We offer customized enterprise licensing, priority onboarding, and dedicated support for large organizations.

Contact Enterprise Sales

See it in action

Before and after — processed entirely on your Mac, in minutes.

Before
Interview_09_original.mov
👤
👤
00:04:231920×1080 · H.264⚠️ PII visible
After
Interview_09_redacted.mov
00:04:231920×1080 · H.264✓ 2 faces redacted
4 min 23 secvideo length
2 facesdetected & redacted
Faster than real-timeon Apple Silicon M2+ (typical)
0 bytessent to any server
⚠️ AI-assisted redaction — always review outputs before sharing. Low-confidence detections are flagged automatically for human review.

Trusted by teams who can't afford mistakes

Early users from law enforcement, healthcare, and legal — in their own words.

“We used to pay a vendor $1,800 a month to process bodycam footage before FOIA releases. Now it runs overnight on a Mac mini in the evidence room. Zero data leaves the building.”
🚔
Records Supervisor
Mid-size municipal police department · Midwest
“Our IRB required that patient video never leave the institution's network. Every cloud vendor we evaluated required a BAA and still stored data on shared infrastructure. Blurr was the only option that was architecturally compliant.”
🏥
Clinical Research Coordinator
Academic Medical Center · Northeast
“Deposition footage for a multi-defendant case — 340 clips, dozens of witnesses to redact. The batch queue processed the entire set over a weekend. The audit log held up in a discovery challenge.”
⚖️
Litigation Technology Manager
Regional law firm · Southeast

Names and identifying details withheld at users' request. Quotes are from real user feedback during early access.

Audit Trail

Every batch leaves a paper trail

Blurr writes a structured JSON audit log for every job — so you can demonstrate exactly what was redacted, with what confidence, using which model version, and when. Built for discovery challenges, FOIA disclosures, and IRB reviews.

  • Source filename, hash, and duration recorded per file
  • Detection type, count, and per-frame confidence scores for every redaction
  • Model version and detection threshold logged for reproducibility
  • Batch start/end timestamps in ISO 8601 for evidentiary timestamping
  • Human-reviewable flag on any frame below your confidence threshold
Read audit log schema →
audit_log_2024-03-15.json
{
  "batch_id": "b7f3a2d1",
  "generated_at": "2024-03-15T09:42:11Z",
  "model_version": "1.0.4",
  "files": [{
    "source": "Cam_042_Main.mp4",
    "sha256": "a3f9...",
    "detections": 47,
    "low_confidence_frames": 2,
    "status": "redacted"
  }]
}

The two questions IT always asks

Forward this to your legal or security team.

🔐

Does my footage ever leave my device?

No. Blurr makes zero network calls during processing. Detection runs entirely on your Mac using Apple Vision and Core ML — the same framework powering Face ID. Your footage never touches a server, ours or anyone else's. You can verify this with any network monitor or firewall log.

🎯

What happens if the detector misses a face?

Blurr flags low-confidence detections so you can review them before the final render. Every output carries a persistent notice — "AI-assisted redaction — always review outputs before sharing" — because automated detection is not 100% and we will not pretend otherwise. The JSON audit log records confidence thresholds per frame so your team knows exactly where to check.

🔑

Can I share one license across multiple machines?

It depends on your plan. The Solo license is tied to a single machine at a time (though you can move it between machines by deactivating). The Team license includes 5 separate machine activations, allowing your whole team to process footage simultaneously. For more than 5 machines, contact our sales team for volume pricing.

Ready to stop paying for cloud redaction?

Join security teams, law firms, and school districts already processing locally.
Solo license from $29/month. Cancel anytime.

Try Free — No account requiredSee pricing →