Ground rules
AI speeds work; it does not replace judgment. Everything we deliver to clients goes through human review. Responsibility always sits with the person who signs off.
Virkum policy
Virkum uses AI deliberately in daily work — as a tool, not a magic solution. This is our stance on how we use it and how we support clients.
AI speeds work; it does not replace judgment. Everything we deliver to clients goes through human review. Responsibility always sits with the person who signs off.
We never send confidential client data into public AI systems without consent. For sensitive work we use paid tiers with enterprise terms that prohibit training on your data. Personally identifiable information is scrubbed before it enters any system.
We use AI in preparation, analysis, drafting and research. That makes work faster and deeper — you get better output in less time, not cheaper work at lower quality.
We do not forward unchecked AI output directly to clients. We do not let AI make decisions about people — hiring, termination or performance reviews. We do not sell “AI” as a product on its own.
A recent Statistics Iceland survey shows that 50% of Icelandic firms use AI but only 14% have a formal usage policy. That gap — AI in daily use without clear guardrails — is one of today’s biggest operational risks.
We help clients bridge it: define policy, scope use cases, and embed it in how the company operates — not as a one-off project, but as part of normal maturity.
Book a free 60-minute call where we review where AI can deliver most value for your company — and where it should not be used.
Book a call→