By · Last updated 2026-04-25

Back to BlogGDPR & Compliance

Token Mapping for GDPR AI Workflows

When customer names are anonymized before AI processing, the AI's response contains anonymized tokens. The final response must contain real names — not.

April 25, 20268 minute read
token mapping AIGDPR customer service AIauto-decryptsession-based anonymizationAI workflow pseudonymization

Token Mapping for GDPR AI Workflows

Updated for 2026

Your team uses AI to draft customer replies. A customer writes in. Their name is anonymized before the AI sees it. The AI drafts a reply with a placeholder. The agent must swap it back manually. At 200 interactions per day, that cost adds up fast.

Session-based token mapping solves this. It restores real names automatically.

The Problem Without Token Mapping

The anonymization step creates a token. "Maria Schmidt" becomes [CUSTOMER_1]. Claude drafts: "Dear [CUSTOMER_1], we apologize for the delay."

The claims handler must now replace [CUSTOMER_1] with "Maria Schmidt" before sending. At scale, this step defeats the purpose of AI assistance. It is repetitive work that does not go away.

How Session Tokens Work

The session stores a lookup table: [CUSTOMER_1] → "Maria Schmidt." When Claude returns its draft, the auto-decrypt layer reads that table and restores the name. The agent sees "Dear Maria Schmidt" — already correct. No manual step. The GDPR protection runs silently.

Why Session Consistency Matters

The token table must be consistent across the full session. If "Maria Schmidt" appears in the initial complaint and again in a follow-up, both must resolve to [CUSTOMER_1]. Without this, Claude may treat them as two different people. Its response becomes incoherent.

One person gets one token per session. Claude can then reason about the conversation correctly.

GDPR Compliance by Design

GDPR Article 4(5) defines pseudonymization as a risk-reduction technique. The EDPB's 2022 guidelines require one thing: the key must be kept apart from the pseudonymized data.

Session token tables meet this rule. The lookup stays in the browser. It never goes to Claude. After the session ends, it is gone. No personal data reaches external servers. The Article 46 transfer question does not arise.

Insurance Claims: A Concrete Example

A German insurer processes customer complaint emails. Each email contains a name, a policy number, and a claim amount.

Before AI processing, the Chrome Extension or MCP Server anonymizes all three fields. Claude sees [CUSTOMER_1], [POLICY_2024-08847], and [AMOUNT_1]. It drafts a reply with those tokens.

The auto-decrypt layer then restores all three fields. The claims handler sees the real name and policy number in the draft. They review and send. No placeholder replacement required.

The GDPR outcome: data sent to Claude's US servers contained no personal data. The customer's real name and policy number stayed in Germany on the handler's browser.

What the Full Loop Requires

Three components must work together for a seamless workflow:

1. Consistent tokens. Each entity gets one token per session. Always the same one.

2. A local lookup table. It lives in the session. It is not sent to the AI.

3. Auto-decrypt on output. The table is applied to the AI draft before the agent sees it.

Without all three, agents replace tokens by hand. With all three, the workflow runs on its own and stays GDPR-compliant.

Conclusion

This approach closes the loop in AI-assisted customer work. Anonymization protects data before it reaches the AI. Auto-decryption puts real names back in the response. Agents see correct names at every step. GDPR compliance holds throughout.

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.