Question 1

What is an AI Interaction Firewall?

Accepted Answer

A security layer between employees and AI models. BRANE inspects every prompt, detects sensitive content, applies policies and decides whether a request is allowed, masked, blocked or processed locally.

Question 2

Is BRANE an appliance or software?

Accepted Answer

BRANE is a hardware appliance on NVIDIA-certified hardware — plug it in, configure, get going. Optionally with controlled cloud access in hybrid mode.

Question 3

Which deployment modes are available?

Accepted Answer

Local-only, local-preferred (recommended for enterprise rollout), cloud-preferred with controls, cloud-only with governance, and air-gap capable.

Question 4

Does BRANE work with ChatGPT, Claude, Gemini and Copilot?

Accepted Answer

Yes. BRANE sits in front: sensitive prompts are masked or processed locally, non-critical ones go to the best model — per policy.

Question 5

How fast is the inspection?

Accepted Answer

Classification in under 50 ms per prompt — with no noticeable delay in the workflow.

Question 6

Does BRANE help with GDPR and the EU AI Act?

Accepted Answer

Yes. Audit logs with content, documented routing decisions and pseudonymization provide the evidence GDPR and EU AI Act Article 50 require.

Question 7

How long does rollout take?

Accepted Answer

BRANE is designed for self-service deployment in under 30 minutes. For larger environments we offer optional onboarding (identity, SIEM, custom policies).

Question 8

Does our data stay in-house?

Accepted Answer

In local mode, no prompt content leaves your environment. In hybrid mode only masked or tokenized content leaves the appliance — never cleartext.

Frequently asked questions.

Which use case fits you?

Let's get in touch.