By · Last updated 2026-04-18

Back to BlogAI Security

3.8 Daily PII Exposures in Support Teams

Every support agent using ChatGPT makes an average of 3.8 sensitive data pastes per day. For a 100-person team, that's 380 GDPR exposure incidents daily.

April 18, 20268 minute read
accidental PII exposuresupport team ChatGPTCyberhaven 3.8 pastesworkflow PII protectionGDPR daily exposure

Daily PII Exposure Math

Cyberhaven research found that enterprise employees make an average of 3.8 sensitive data pastes into ChatGPT per user per day. For a 100-person support team, that is 380 instances of customer records entering ChatGPT every day.

Each instance can be a GDPR data minimisation breach under Article 5(1)(c). That article requires personal information to be "adequate, relevant and limited to what is necessary."

These are not rogue employees ignoring policy. The 3.8 figure reflects normal work. Agents copy customer emails to draft replies. They paste complaint text to get empathetic suggestions. They include account details to get context-aware answers. Each paste is a valid productivity step that happens to carry PII along.

Behavior Training Does Not Fix This

A 2024 EU audit found that 63% of ChatGPT user data contained personal identifiable information. Only 22% of users knew they could opt out via the tool's settings. Most content pasted into an AI assistant contains PII. Most users are unaware of the controls. The result is daily exposure at scale.

Policy training runs into a basic problem. The copy-paste habit is decades old. Users have been copying and pasting text since their first day at a computer. Plugging an AI chat tool in as a paste target adds a new destination. It does not change the habit.

A "do not paste customer PII into the AI assistant" policy asks agents to insert a classification step — "does this text contain PII?" — into a habitual action that has no natural pause. Training effects wear off. The cumulative result of 380 daily paste decisions is a compliance risk that policy alone cannot hold.

Where Technical Controls Work

The fix operates at the paste action itself. A browser extension intercepts clipboard content the moment the agent presses paste, before the text reaches the input field. The agent sees a preview modal. It shows what was detected and what will be anonymised before the text is sent.

This is not a blocking control. Agents can proceed, override, or stop. It is a transparency step. It adds one moment of visibility to an otherwise automatic action.

Consider a German e-commerce support team lead drafting replies to customer complaints. The workflow stays the same: copy the complaint, paste into ChatGPT, generate a reply. The extension adds a two-second check. The agent sees that names, addresses, and order numbers were detected. The agent clicks proceed. The tool receives the anonymised version. The compliance breach does not occur.

Our GDPR compliance guide covers the legal basis for these controls. See also our AI policy vs. technical controls comparison and the browser DLP for ChatGPT guide for implementation detail.

Sources

Ready to protect your data?

Start anonymizing PII with 285+ entity types across 48 languages.

About this page

We update this page when our platform or the law changes.

Read our founder note for how we work.

Each change shows up in the timestamp at the top.

Related reading

We follow these rules

  • GDPR (EU 2016/679).
  • ISO/IEC 27001:2022.
  • NIS2 (EU 2022/2555).
  • HIPAA safe harbor under 45 CFR § 164.514(b)(2).

Our promise

We do not sell your data.

We do not train models on your text.

We store your files in Germany.

You can delete your account at any time.

You own your work.

Where we run

Our servers live in Falkenstein, Germany.

We use Hetzner. They hold ISO 27001 certification.

All data stays in the EU.

Backups run every day.

Need help?

Email support@anonym.legal.

We reply within one business day.

How we test

We run a full check suite on every release.

Each surface gets its own sweep script and report.

Human reviewers spot-check the output each week.

We track recall and precision on a labelled set.

Bad runs block the deploy.

What we never do

  • We never sell your information to third parties.
  • We never train models on what you upload.
  • We never keep your work after you delete it.
  • We never share keys with any outside firm.
  • We never run ads inside the product.

Plans in plain words

We sell credits, not seats.

One credit covers one short job.

Long jobs use a few credits each.

You can top up at any time.

Unused credits roll over each month.

Read the plans page for current rates.

Who built this

A small team of engineers and lawyers built this.

We ship from Europe and work in the open.

Our founder note spells out why we started.

Where to start

How the parts fit

A browser add-on cleans text inside Chrome.

A Word plug-in handles drafts in Office.

A small desktop tool works on whole folders.

An agent protocol link feeds large models safely.

All four share one core engine and one rule set.

Words from our team

We started this work after a lunch about cookies.

One friend kept getting odd ads on her phone.

We asked why a court file leaked through a draft.

We sketched the first build on a napkin that week.

By month three we had a tiny demo for a friend.

She used it on her first case the next day.

Common questions we hear

Can the tool read scanned PDFs? Yes, with OCR.

Does it work on long files? Yes, in small chunks.

Can I roll my own rule set? Yes, save it as a preset.

Does it run offline? The desktop build runs offline.

Do you keep my files? No, the cloud build wipes after each run.

Will it learn from my work? No, we never train on inputs.

A short tour of the workflow

Upload a file or paste a snippet of prose.

Pick the entities you want gone from the draft.

Choose a method: replace, mask, hash, encrypt, or redact.

Press run and watch the side panel show each hit.

Skim the result and tweak any rule that misfired.

Save the cleaned file or send it to a teammate.