xFleet: Threshold-Unlock AI Model Deployment

The Problem

Deploying AI models to edge devices creates thousands of copies that can be extracted, reverse-engineered, or tampered with.

Edge AI is moving 80% of inference to local devices, but each device holds a complete copy of the model weights. A single device compromise exposes the entire model — months of training and millions of dollars in IP.

Current protection relies on obfuscation and TEEs, but obfuscation is reversible and TEE availability is limited on consumer hardware.

The Old Way

The PRIVATE.ME Solution

xFleet splits model weights via XorIDA across fleet devices. No single device holds a complete model. Inference requires threshold reconstruction from K cooperating devices.

Model weights are split into K-of-N XorIDA shares at deployment time. Each device receives one share. Inference requires K devices to cooperate and reconstruct the relevant layers. xBoot ensures integrity verification before reconstruction.

Deployment uses HMAC-signed manifests with per-chunk SHA-256 hashes. Tampered devices are detected and excluded from the threshold group automatically.

The New Way

How It Works

xFleet combines XorIDA model splitting with xBoot integrity verification and threshold reconstruction for secure fleet-wide AI deployment.

Use Cases

⚡ Fast Onboarding

Traditional fleet management requires manual custodian setup, share distribution configuration, and attestation key provisioning. xFleet collapses this to 15 seconds with zero-click accept, 90 seconds with one-line CLI, and 10 minutes with deploy buttons.

Speed Tiers

Quick Start: Zero-Click

# 1. Set invite code (from fleet operator onboarding email)
export XFLEET_INVITE_CODE='XFL-abc123'

# 2. Register your first vehicle (auto-accepts invite)
npx @private.me/xfleet register \
  --vin '1HGBH41JXMN109186' \
  --make 'Honda' \
  --model 'Accord' \
  --year 2021

CLI Setup (90 seconds)

# Install CLI globally
npm install -g @private.me/xfleet

# Initialize (generates custodian DIDs, saves to .env)
xfleet init

# Register your first vehicle
xfleet register \
  --vin '1HGBH41JXMN109186' \
  --make 'Honda' \
  --model 'Accord' \
  --year 2021

# Output:
# ✅ Vehicle registered (2-of-3)
# ✅ Shares distributed to custodians
# ✅ Odometer attestation enabled
# ✅ Ready to track telemetry

Deploy Button Setup (10 minutes)

Click one button to provision telemetry server + custodian share storage on Vercel/Netlify/Railway:

Integration

import { FleetDeploy } from '@private.me/fleetdeploy';

const deployment = await FleetDeploy.create({
  modelPath: './model-weights.bin',
  fleet: deviceDids,
  threshold: { k: 3, n: 5 }
});
await deployment.distribute();

Security Properties

Verifiable Data Protection

Every operation in this ACI produces a verifiable audit trail via xProve. HMAC-chained integrity proofs let auditors confirm that data was split, stored, and reconstructed correctly — without accessing the data itself.

Ready to deploy xFleet?

Talk to Ren, our AI sales engineer, or book a live demo with our team.

Ship Proofs, Not Source

xFleet generates cryptographic proofs of correct execution without exposing proprietary algorithms. Verify integrity using zero-knowledge proofs — no source code required.

Use Cases