Cranium

Primary category: ai-spm. Also spans ai-governance-platform.

One-liner — An enterprise AI security + governance platform that inventories your AI/ML systems, assesses their risk and compliance posture, and documents trust for internal and third-party AI.

What it does

Cranium helps an organization map, monitor, and manage its AI/ML estate against adversarial threats “without interrupting how teams train, test, and deploy.” It builds an AI inventory, runs security-risk and exposure assessments, and ties findings to compliance frameworks (NIST AI RMF, EU AI Act, etc.). A distinctive piece is AI Cards / third-party AI trust — documenting and exchanging the security posture of AI systems with customers and across the supply chain. So it straddles posture (AI-SPM) and governance/compliance evidence.

Where it sits in the stack

Primary ai-spm (discover + assess + monitor AI assets), with heavy overlap into ai-governance-platform (compliance mapping, model documentation, third-party AI trust). Layer: model/prompt, reaching up into the governance layer. It reduces risk indirectly — by finding weak/untrusted AI surfaces and risky data handling — rather than blocking traffic inline. More a posture/governance tool than a runtime control.

Deployment & architecture

SaaS platform with API connectors to discover and assess AI/ML systems across the enterprise; produces reports, AI Cards, and compliance artifacts. Not an inline proxy or runtime firewall in its core form — though the 2026 Aiceberg acquisition adds agentic-AI runtime/risk capabilities (see below).

Positioning & differentiators

Known for the KPMG pedigree (incubated in KPMG Studio; KPMG is investor and go-to-market channel) and for the AI-trust / AI Cards / supply-chain angle — quantifying and sharing third-party AI risk, which fits TPRM and audit workflows more than a pure security tool. Differs from posture-plus-runtime players like noma-security and zenity (more runtime/agent focus) and from governance-first platforms like credo-ai, holistic-ai, and ibm-watsonx-governance (policy/model-risk first). Cranium tries to bridge security posture and governance evidence in one place.

Ownership, funding & M&A

Independent, VC-backed. Founded 2023 by Jonathan Dambrot (CEO), Daniel Christman, Felix Knoll, and Paul Spicer; the first technology spinout from KPMG Studio. Series A $25M (Oct 2023) led by Titanium Ventures with KPMG LLP and SYN Ventures; ~$32M total.

M&A (verified): On 2026-05-21 Cranium acquired Aiceberg, an agentic-AI security and risk-management company, to extend toward end-to-end coverage including AI agents. Cranium is the acquirer and remains independent — not itself acquired. (No “Cranium acquired” flag existed in the seed; confirmed direction against the BusinessWire release.) Ownership confidence: high.

CTO / hedge-fund lens

Day-2, and skews toward the governance/compliance buyer rather than the SOC. If your driver is SR 11-7 model-risk evidence, EU AI Act / NIST AI RMF documentation, and third-party AI vendor risk (which AI is in your supply chain and is it trustworthy), Cranium’s AI-Cards/trust framing is its strongest hedge-fund hook. If your driver is blocking prompt injection at runtime, look at runtime players instead. Mid-to-large regulated shops get the most out of it.

Competitors / alternatives

noma-security, zenity, reco, credo-ai, holistic-ai, modelop, ibm-watsonx-governance, onetrust.

Open questions / to verify

  • What Aiceberg adds in practice (agent runtime vs risk scoring) and how it changes deployment.
  • Whether Cranium offers any inline/runtime enforcement post-acquisition or stays posture/governance-only.
  • Current total funding after 2023 (any undisclosed rounds before the Aiceberg deal).

Sources

History

  • [2026-06-28] Stub created from seed registry.
  • [2026-06-28] Researched; established independent KPMG Studio spinout (2023), Series A $25M (2023-10, Titanium Ventures), Short Hills NJ. Verified Cranium is the ACQUIRER of Aiceberg (2026-05-21), not acquired. Set ownership_confidence high.