Skip to content
Playbook
IntermediateHRPeopleOps

Shadow AI: what your team is already using and how to surface it

Banning AI doesn't stop AI. It just pushes it onto personal devices and personal accounts where you have zero visibility.

8 min read
On this page
60-Second Summary
  • If your policy is 'no AI', 60-80% of knowledge workers are still using it — on phones, on home laptops, on personal accounts.
  • Amnesty + enterprise alternative beats enforcement. Always.
  • The risk isn't 'AI exists', it's 'we can't tell what data left the building'.
  • Measure success by drop in personal-account usage, not by 'caught' incidents.

Shadow AI is the natural consequence of a fast tool plus a slow procurement process. Engineers won't wait three months for a Copilot license — they'll paste code into a personal account today. Marketers won't wait either. Your job isn't to catch them; it's to make the sanctioned path faster than the shadow one.

Why shadow AI happens

  1. The official tool is slower, worse, or not yet approved.
  2. Approval requires a security review that takes longer than the project.
  3. Nobody told the user there was an approved alternative.
  4. The user knows but doesn't trust the approved tool's logging.

The 30-day amnesty program

Week-by-week
  1. 1
    Week 1
    Announce: 'For the next 30 days, declare what AI you use — no consequences. After day 30, undeclared use is treated as a policy violation.'
  2. 2
    Week 2
    All-hands + manager 1:1s. Each team lists tools, use cases, what data they put in. Stand up a Slack channel for questions.
  3. 3
    Week 3
    Procurement fast-track for the top 5 tools surfaced. SOC 2 + DPA review in 5 working days, not 30.
  4. 4
    Week 4
    Publish the approved list, the policy, and the 'how to switch' guide. Run a clinic to migrate accounts.

Provide the enterprise alternative

Every shadow tool that survives the amnesty should have a sanctioned alternative within 60 days. If you can't fund alternatives, you don't have an AI policy — you have an AI ban, and bans don't work.

What to monitor (and what not to)

Visibility without surveillance
Do monitor
  • Outbound traffic to known AI domains from corp devices
  • DLP rules for source code + customer data
  • Enterprise account usage trends
  • Quarterly self-attest in the AI register
Don't monitor
  • Keystroke loggers
  • Slack/email content scans for 'prompt' or 'ChatGPT'
  • Personal device traffic
  • Individual prompt history without cause
Written by Pawan Joshi.Sources cited inline.
First published 16 Jun 2026See site changelog →