Month	Capital	Southeast	Southwest
Jan	4,120	2,210	1,080
Feb	4,380	2,340	1,120
Mar	4,510	2,280	1,150

Slide	Content
1	Cover
2	Key points
3	Shipments

Solutions · Financial services

Control and record AI use
in line with Korean financial security guidelines

Syncgate pseudonymizes personal (credit) information before sending it to a model, allows only approved models, and records every decision and administrator action in a sealed audit log. Where network separation is required, it is installed on an air-gapped network and connects only to in-house models.

Contact sales Korean financial security guidelines

Contact details and card numbers replaced with pseudonyms before they are sent to the modelReal product screen

Financial-sector requirements

Key points financial institutions check
and the features that address them

Personal (credit) information

Personal (credit) information detection and pseudonymization

Detects 13 data types, including resident registration numbers, account numbers, card numbers and driver's license numbers, in input and output, and blocks or masks them. Only pseudonyms are sent to the model, and the original values are restored when the answer is shown to the user. It can also warn at the input stage and ask the user to record a reason.

FSC guideline check 7 · FSS AI RMF privacy · NIS guidebook M13Guardrails

Prompt injection

Layered defense against prompt injection and jailbreaks

On top of blocked-word and regular-expression rules, a guardrail model that judges each sentence in its context runs a second check. With “Pause after repeated violations” turned on, a user is warned after 5 violations and blocked for 30 minutes after 10. These defaults match the compliance example in the guidelines.

FSC guideline checks 6, 9 · NIS guidebook M13, M15Guardrails

Model control

Approved models only

Users and company systems do not call models directly. Their requests are sent through the gateway. Only providers and models an administrator has approved are allowed, per group or per user, and request and token limits are set for each user. On an air-gapped network, only self-hosted models are used.

NIS guidebook M17 · FSC guideline check 12 · N2SF Model 2Model gateway

Audit

Log retention and tamper evidence

Records requests, guardrail decisions and administrator activity. The audit log is kept indefinitely by default, checked for tampering with a hash chain, and sent to your SIEM over syslog in real time.

National Cybersecurity Basic Guidelines Art. 55 · NIS guidebook M09 · Compensating controls for financial network separationAdministration and audit

Access control

MFA, least privilege and human approval

SSO, SCIM and two-step verification with TOTP or passkeys are supported, and least privilege is applied per group. Two-step verification can be made mandatory for administrators. Tools that change external systems run only after human approval, and the chat input shows the security warning your organization sets.

FSC guideline check 24 · National Cybersecurity Basic Guidelines Art. 76 · NIS guidebook M05 · M30 · A-M12Permissions and share review

Network separation

Verified air-gapped installation

On a network with external internet access blocked, we verified installation, first-account sign-in, connecting an in-house model, document upload and indexing, search and chat. You need only the installation image files, a license file and the address of your in-house model.

Installation and operations

See all six guideline items

Guardrails

Sensitive data is protected
before it reaches the model

Rule-based detection, AI evaluation and prompt injection detection inspect every input and output. Depending on policy, data is blocked or pseudonymized, and every verdict is recorded in the guardrail logs.

Chat inputTry editing it

Chat input

Action

What the model receives

Where checks run

Detection runs on every request at each enforcement point. Hover over or select an item in the diagram to see its decision.

Sample

13 enforcement points

Chat inputModel outputAttachmentsText in imagesTool resultsMemoryDB extractionFilesScheduled tasksAPICode gatewayAsk the manual

13personal data types6actionsConfidential document similarity detection

FAQ

Questions financial institutions often ask

Is customer information sent to the model as is?

Guardrails can prevent that. They detect 13 types of personal (credit) information and either block them or replace them with pseudonyms before the request is sent to the model. Personal data columns in databases are masked by column policies.

Can it be used where network separation is required?

Yes. Install it on an air-gapped network and connect only in-house models. Everything from installation to chat has been verified in an environment with no internet connection.

Can the results be used for audits?

The audit log and conversation logs show who requested which task and what the result was, and completion checks recompute the figures in the result files to confirm them. The results themselves are drafts, however, and the final decision rests with the person in charge.

Related features

Knowledge search and cited answersAnswers with a cited source for every sentence Database queriesQueries business databases in natural language and answers with tables and charts Document connectorsBrings in and indexes documents from 38 document sources Security and complianceKorean financial and public-sector guidelines

Examples of work done with each harness templateSee workflow demos

See it with
your own documents

In a consultation, we connect a few of your documents so you can check the answer quality for yourself. Send us your questions and our team will get back to you.

Contact sales Request security documentation

Control and record AI usein line with Korean financial security guidelines

Key points financial institutions checkand the features that address them