Concepts

Siper concepts: what happens in the masking step, how to read the placeholder format, what fail-closed behavior does, and where your data is processed.

Every request sent through a virtual key with masking enabled passes through a masking step first. This page explains what happens in that step: the format of the placeholders, how the Gateway behaves when masking cannot be performed, and where your data is processed.

How does masking work?

Text arriving at the Gateway is scanned before it is forwarded to the model. Each detected piece of personal data is replaced with a placeholder in the form <KATEGORİ_N>: the first person name in the text becomes <KİŞİ_1>, the second <KİŞİ_2>. For detected values, the model sees these placeholders instead of the original values; the placeholders are preserved as the response comes back.

Detection covers 26 categories, including names, TCKN (Turkish national ID number), phone numbers, and addresses. For the full list and synthetic examples, see Categories.

Behind this accuracy is data: we built a large Turkish training dataset that covers KVKK’s special categories (including health, religion, political opinion, trade-union membership, and criminal convictions) and trained the model on it over months of work. We re-measure new releases under the same test protocol and share comparative results on the benchmark page. Recall measured 84.6% in the July 2026 evaluation; no automated detector can guarantee that it will find every piece of personal data.

Placeholder format and the language distinction

Direct /mask and /mask/batch responses use raw category names such as <PERSON_1>, <PHONE_MOBILE_1>, and <TCKN_1>. The OpenAI-compatible Gateway and the management panel display the same categories with Turkish labels. Each new value in the same category increments the number by one.

In integration code, follow the contract of the surface you call: direct masking endpoints return raw category names, while the Gateway returns Turkish labels.

What happens if masking fails?

The request stops. If the masking service cannot process a request, the Gateway does not forward that request to the external model and responds with a 502 status code; there is no “send it unmasked anyway” option. When you receive a 502, retry the request later. This behavior is called fail-closed.

Data sovereignty

Masking runs before the model call, on infrastructure hosted in Turkey. With masking enabled, detected personal data is converted to placeholders in Turkey and those masked values are sent to the external language model.

For retention, audit trail, and access topics, see Security.