Personal phone with no MAM policy leaks corporate Outlook cache
Your scan flagged 38 personal devices accessing corporate mail without an App Protection Policy — this scenario walks through what happens when one is stolen.
- NIST CSF
- Protect (PR.PT)
- Generated at
- May 7, 2026, 14:32 UTC
- Source job
- demo-tune-0419
- Source checksum
- sha256:b39c…2af0
Run drill — live (demo)
Try the live drill runner — timer, inject Drop buttons, response capture, scoring rubric, the lock flow, AI Coaching feedback simulation, and the spectator URL handoff for attendees. No tokens consumed; no audit record produced. The lock will synthesize a real-looking session checksum (browser SHA-256) so you can see the locked-evidence view.
Open demo runnerBundle versions & edit history
Draft · editableThis drill bundle is editable until the first session is created. Each save appends a new version (v1, v2, …) with a SHA-256 captured in the audit chain — auditors can see exactly what changed, when, and by whom. The first Run drill click freezes the bundle: whatever version is active at that moment becomes the locked auditor-evidence source for every session that follows.
Sessions started right now will run against v1. The bundle freezes at v1 the first time someone clicks Run drill — make any edits before then.
Personal phone with no MAM policy leaks corporate Outlook cache
Veri-Tune · v1.0.0Your scan flagged 38 personal devices accessing corporate mail without an App Protection Policy — this scenario walks through what happens when one is stolen.
Protect (PR.PT)
60 minutes
5 timed
6 criteria
On a Friday evening, a senior engineer's personal Android phone is stolen at a transit station. The phone has Outlook for Android installed, signed in to corporate identity, with 14 days of cached mail including draft RFP responses and a board-deck PDF. The phone is unlocked when stolen (the engineer was actively using it). Your scan showed the device is enrolled in compliance reporting but has NO App Protection Policy applied — meaning the corporate data on the device is not encrypted at rest by the corporate-controlled key, and a remote wipe would only work if the device is online and we can locate it.
Threat actor: Opportunistic theft, low-skill. Phone may be wiped and resold OR data may be exfiltrated if the thief is more sophisticated than typical.
Attack chain
- 1Initial access: Phone stolen unlocked at transit station. Outlook is the front-most app.
- 2Data discovery: Thief (or buyer) explores the Outlook cache. 14 days of mail visible without re-authentication because session is active.
- 3Exfiltration: Cached attachments (board deck PDF, RFP drafts) downloadable via 'Save to Photos' → uploaded to attacker cloud storage.
- 4Persistence (optional): If thief is sophisticated: install a forwarding rule from the device, gaining persistent visibility into mail until detected.
Affected assets
- Senior engineer's corporate mailbox (90 days mail history accessible via cache + reauth)
- Board deck PDF (revenue projections + planned acquisitions)
- RFP response drafts (competitive intelligence)
- Cached calendar with attendee names + meeting subjects
Linked scan findings
| Control ID | Severity | Finding |
|---|---|---|
| INT-AP-001 | High | App Protection Policy not applied to BYOD Android devices accessing corporate Outlook |
| INT-CA-014 | High | Conditional Access does not require app-protection state for Outlook mobile |
| INT-CMP-007 | Medium | Device compliance reporting enabled but no policy enforcement |
Generated from Veri-Tune scan demo-tune-0419 on 2026-05-07. This is facilitator material — verify scenario specifics against your tenant before use. Veri-Tech does not warrant scenario fitness for any specific audit framework; pair with the source scan job (which IS auditor evidence) and your own IR plan.
AI generation provenance
- Model
- claude-haiku-4-5-20251001
- Template version
- v1.0.0
- Generated at
- May 7, 2026, 14:32 UTC
- Org-shape snapshot
- midmarket · 7 roles frozen at generation time
Auditors verify AI-generation lineage by reading _manifest.json in the source bundle (full token + cache accounting, generation timestamps, SHA-256 cross-references). The auditor ZIP carries it verbatim.
Auditor-grade artifacts
The three audience-tailored downloads below are demo replicas of what a locked session would produce — full Team Debrief PDF, board-packet Executive Brief PDF, and the Auditor ZIP with bundle audit chain + SHA-256 cross-references. The locked session backing these demo artifacts is pre-populated with realistic responses + scores so the PDFs render against meaningful content. Production artifacts ship with a 6-year WORM retention contract on a real Vault tenant; these are clearly watermarked as demo.
Edit history
Draft (editable)Active version: v1. No sessions have been created yet — the bundle is still editable. The chain freezes at the first session creation.
- v1 Facilitator save·May 7, 2026, 14:42 UTC
- By
- demo.facilitator@veri-tech.net
- Bundle hash
- sha256:b9d4c8e3f2a1…
“Tightened the T+25 min inject + added missing vendor-coordination workload to the IR plan”
Fields changed (1)- injects[1].content
- v0 AI baseline·May 7, 2026, 14:32 UTC
- By
- claude-haiku-4-5-20251001
- Bundle hash
- sha256:a8c3b7f2e1d9…
Each version’s SHA-256 is captured in the IrTabletopBundleEdited / IrTabletopBundleFrozen audit events (App Insights, 6yr retention) and exported into the auditor ZIP’s manifest.json.
Demo mode: editable-drafts saves never touch a server, no Anthropic tokens are consumed, downloads carry a “DEMO ARTIFACT” watermark in their manifest disclaimer. In production: each save POSTs to /api/ir-tabletop-bundles/{drillId}/versions, computes a real server-side diff + SHA-256, appends to the WORM-protected audit chain, and emits an IrTabletopBundleEdited App Insights event with 6-year retention.
