The News

What was announced?

Cranium AI has acquired Aiceberg, an agentic AI security and runtime protection company. The two complementary AI security companies are now unified under a single brand, Cranium AI. Together, we form a more complete platform for detecting, testing, and securing AI across the enterprise.

Why did this happen?

As organizations increasingly adopt agentic AI and autonomous workflows, the need to connect governance, runtime security, and operational oversight continues to grow. Cranium has built deep capabilities in AI governance, risk management, and third-party AI oversight across the build and procurement stages of the AI lifecycle. Aiceberg brings a real-time AI firewall and runtime enforcement layer, with advanced protection against prompt injection, sensitive data exposure, and agent policy violations. Together, Cranium AI delivers end-to-end lifecycle coverage: from the AI you build and procure, to the agents you deploy, to the interactions they generate in real time.

What This Means for You

Will anything change about my current product or service?

No, not right now. Your existing products, contracts, support relationships, and points of contact remain exactly as they are. There are no immediate changes to your experience.

Will my pricing or contract terms change?

No. This announcement has no impact on your current contract or pricing terms.

Who do I contact if I have questions or need support?

Your existing account team and support contacts remain in place. For general questions, or if you’re unsure who to contact, you can also reach us at [email protected].

Looking Ahead

Will Aiceberg’s capabilities be added to the Cranium AI platform?

Yes. We’re working through the integration roadmap now and will share specifics well before anything affects your experience.

What are the most exciting things the combined company unlocks?

A few capabilities we’re especially excited about:

• Real-time AI runtime protection: real-time monitoring and enforcement across LLM and agentic interactions, built on deterministic (non-LLM) models, so you’re not using AI to protect AI
• Deeper agentic security: connecting Aiceberg’s agent action controls with Cranium’s AgentSensor® and governance capabilities
• Expanded continuous compliance: tying runtime audit logs into Cranium’s compliance reporting across frameworks such as NIST AI RMF, ISO 42001, EU AI Act, and DORA
• Unified AI risk visibility: a single platform that spans governance, testing, detection, and response

Where can I learn more?

We’ll be hosting a customer briefing in the coming weeks to walk through our vision for Cranium AI and answer your questions directly. In the meantime, visit cranium.ai or reach out to your account contact.