Frequently Asked Questions

BlackGrid is a UK-built sovereign AI defence system that stops fraud, impersonation, and AI-driven attacks before trust, identity, or damage occurs.

Because security was built to protect systems after identity appears.

AI made identity cheap to fake.

BlackGrid was built for the layer before identity is trusted — which didn’t need to exist until now.

Because it would break their existing products.

BlackGrid doesn’t depend on identity, endpoints, or networks — theirs do.

No. We never read content, messages, or files.

We analyse how a signal behaves, not what it contains.

We verify location truth, signal authenticity, and behaviour consistency first.

If those don’t align, identity cannot be trusted — even if it looks valid.

We stop attacks before login, before approval, and before transactions — where traditional tools are blind.

No. Every decision includes clear reason codes — such as location mismatch, signal replay, or behavioural anomaly.

Lightweight deployment.

Observe-only first.

TrustScore builds safely.

Enforcement only when the organisation is ready.

No. BlackGrid sits before them, feeding safer decisions into what you already use.

We don’t block blindly.

Decisions are learned safely over time before enforcement is enabled.

They can try once.

BlackGrid safely detonates attacks, learns them, and closes the gap continuously.

Yes. No PII, no surveillance, explainable decisions, UK-built, UK-controlled.

It’s a pre-identity safety layer that integrates with existing systems — not a replacement category.

Because AI has broken identity — and reacting after damage is no longer acceptable.

Because we operate in the blind spot before harm occurs, without invading privacy.