Text Link
1/5

Desired Service

PROJECT DETAILS

2/5

BUDGET

3/5

TIMEFRAME

4/5

CONTACT DETAILS

We use your details solely to process your enquiry. Details in our privacy policy.

5/5

THANK YOU

Our AI agent was quick: the reply to your initial enquiry is already in your inbox.

ERROR

THANK YOU!

We will get back to you shortly.

Data protection · AI systems in the company

GDPR and AI, before the first prompt

The moment a language model sees customer data or personnel files, what the GDPR asks of you changes. We build AI systems so that your data protection officer and works council can read the log. The result alone will not satisfy either of them.

DPA checked under Art. 28DPIA under Art. 35Deletion down to the indexEU or your own hardware
Art. 30
Record of processing
rewritten when AI comes in
72 h
Breach notification
Art. 33, from awareness
6
Documents handed over
from the record to the deletion job
0 EUR
First review
in the first call, even if nothing is missing
GDPR (General Data Protection Regulation)
The GDPR is Regulation (EU) 2016/679, the General Data Protection Regulation. It applies to any organisation processing personal data in the EU, and it keeps applying unchanged when an AI system takes over the processing. The law stays the same. The path the data takes changes, and that path has to be written down again.
01 Starting point

The model is a new recipient. That is the whole difference.

A clerk who summarises customer emails with ChatGPT has, in GDPR terms, just transferred data to a processor in the United States. No contract, no entry in the record, and nobody knows how long it is kept.

That is how it starts in most companies: browser tabs in three departments, long before anyone calls it an AI project. The data protection officer hears about it at the annual review.

The regulation itself has not moved. Legal basis, purpose limitation, processor contracts, data subject rights: all known since May 2018. What is missing is the translation to prompt, log and vector index. That is what we deliver as the technical partner, with your data protection officer at the table. In the knowledge agent project index and search run on EU servers for that reason, and the speech recognition in the meeting project runs on our own hardware.

48.83 % cite data protection
Among EU enterprises that considered AI and decided against it, 48.83 % name concerns about data protection and privacy as a reason. Settle the record, the contract and the deletion path first, and that reason is gone.
Source: Eurostat, Use of artificial intelligence in enterprises, 2024 survey, published December 2025. ec.europa.eu/eurostat
02 Obligations

Six places where the GDPR bites when AI comes in

Each one has an article number and a technical counterpart. We take care of the second half and hand the first half its material.

01

Legal basis

Before a model sees data, the processing needs a basis under Art. 6: contract, legitimate interest or consent. A new purpose, say summarising instead of handling, needs a fresh check under Art. 5(1)(b).

Art. 6Art. 5Purpose limitation
02

Data processing agreement

The model provider is a processor. Art. 28 requires a contract with fixed content plus the list of sub-processors. Without it every API call is a transfer with no basis.

Art. 28DPASub-processors
03

Impact assessment

Art. 35 calls for a DPIA where a high risk is likely: employee scoring, applicant screening, large data merges. Where none is needed we record that too, with reasons.

Art. 35DPIAReasoning
04

Information

Customers and employees must know that an AI system reads along, for what and for how long. Art. 13 and 14 require this before the first processing, in plain language. We supply the text for the privacy notice.

Art. 13Art. 14Transparency
05

Data subject rights

Access, rectification, erasure, objection, Art. 15 to 21. Plus Art. 22 where the system decides about people. Technically this means every person must be findable in prompts, logs and index.

Art. 15 to 22AccessObjection
06

Deletion concept

Storage limitation under Art. 5(1)(e) and erasure under Art. 17 also cover embeddings, caches and backups. A deletion job that only reaches the database does not meet them.

Art. 17RetentionIndex
03 Data flows

Where the data goes, and what that triggers legally

The choice of where the model runs decides the contract, the third country question and the effort. The table shows the four usual routes.

Where the model runsLegal basis and contractThird country transferOur recommendation
Cloud API of a US providerDPA under Art. 28 mandatory, transfer secured under Art. 45 or 46 (adequacy decision or standard contractual clauses)Yes, Chapter V applies, even with an EU region at the providerOnly with pseudonymised input, training excluded by contract, retention switched off
EU provider or EU regionDPA under Art. 28, read the sub-processor listNo, as long as support and backups stay in the EUDefault for customer data, keep a model switch open
Own hardware, local modelNo processor for the model, record and deletion concept still requiredNoFor personnel files, health data and anything the works council sees
Free accounts in the browserNo DPA possible, consumer terms, training usually onYes, uncontrolledSwitch off and replace with a company account under a DPA
04 Engineering

What the code contributes

Art. 25 requires data protection by design, Art. 32 appropriate security. Both are design decisions, not appendices. Four of them are in every system we hand over.

In Germany the works council has a co-determination right under section 87(1) no. 6 of the Works Constitution Act for systems that can capture employee behaviour or performance. We build the log so that it can say yes.

Pseudonymisation
Names, customer numbers and addresses are replaced before the prompt leaves the building and put back after the answer. The model sees placeholders.
Logging
Every model call is logged with time, purpose and data category. The content is stored only when the purpose needs it.
Access rights
The agent sees what the person using it sees. No shared account with full access to the file share.
Retention periods
Prompts, answers and caches carry an expiry date. The deletion job reaches the vector index and the backups, and it is logged.
05 Approach

Three steps, six documents

We start with the processing that is already running. That is where a supervisory authority asks first.

Step 1

Inventory

Which department uses which tool, with which data, under which account. Including the tools nobody ever purchased.

InterviewsAccount listData categories
Step 2

Paperwork

Entry in the record under Art. 30, contract review under Art. 28, DPIA decision with reasons, text for the privacy notice under Art. 13.

Art. 30Art. 28Art. 35
Step 3

Implementation

Pseudonymisation, logging, access model and deletion job in the code. Handover to the data protection officer and the works council with a technical description both can read.

CodeDeletion jobHandover

The six documents: record entry, DPIA decision, DPA checklist per provider, deletion concept, technical description, information text for customers and employees. All in a format you can keep up to date yourself six months later.

This page is not legal advice. We deliver technology and paperwork. The legal assessment is made by your data protection officer or your law firm, and we are happy to work with both.

06 Boundary

GDPR and EU AI Act: two laws, one register

The EU AI Act looks at the system, the GDPR at the data inside it. An internal chatbot often falls under both.

On one side the labelling duty under Art. 50 of the AI Act, on the other the record, the contract and the retention period. Handle the two laws separately and you inventory the same bot twice.

We keep both in one table. What the AI Act asks of deployers is on the AI transparency page and in the EU AI Act documentation.

07 Common questions

GDPR and AI: common questions

Can staff paste customer data into ChatGPT?+
With a private or free account: no. There is no data processing agreement, the input may be used for training, and nobody can tell you how long it is kept. With a company account, a DPA under Art. 28 GDPR, training switched off and pseudonymised input, the picture changes. That is the setup we build.
Do we need a data protection impact assessment?+
Not for every AI system. Art. 35 GDPR requires a DPIA where a high risk to the rights of individuals is likely: scoring employees, screening applicants, merging large data sets. A chatbot over product manuals usually does not qualify. We write down the decision against a DPIA with its reasons. That is the first thing a supervisory authority asks for.
What has to be in the contract with the model provider?+
The items in Art. 28(3) GDPR: subject matter and duration, type of data, processing on documented instructions, confidentiality, sub-processors, help with data subject requests, deletion or return at the end, and audit rights. The big providers have these contracts ready. You still have to read them, above all the sub-processor list and the clause on how long inputs are retained.
How long may prompts and answers be stored?+
As long as the purpose requires, Art. 5(1)(e) GDPR. For a debugging log that is days to a few weeks; for a ticket system, as long as the ticket. What matters is that somebody sets the period and a deletion job enforces it, including in the provider’s cache. Most providers offer a zero retention setting for exactly this.
What about employee data and the works council?+
Once an AI system logs what employees type, it can capture their behaviour or performance. In Germany the works council then has a co-determination right under section 87(1) no. 6 of the Works Constitution Act. We recommend involving it before the first log entry and designing the log for debugging, not for performance monitoring.
Is an EU region at the cloud provider enough?+
For the place of processing, usually yes. The review should also cover where support sits, where backups live, which sub-processors have access, and whether the provider itself is subject to third country law. That is in the contract. The product page rarely says.
What does Kuroko Labs do here, and what not?+
We deliver the technology and the paperwork: the record entry, the DPIA decision, a DPA checklist per provider, the deletion concept, pseudonymisation and logging in the code. The legal assessment stays with your data protection officer or your law firm. We are not legal advisers and do not present ourselves as such.
08 Read on

Related topics

Next step

Tell us which department is already using AI today.

Thirty minutes are enough to say whether the record and the contracts need catching up. If they do not, we will say that too.

DEJPEN