Question 1

What is the Bollwerk AI-Threat Readiness Assessment?

Accepted Answer

A free assessment that evaluates whether an organisation's controls can withstand AI-enabled fraud, synthetic media, prompt injection, agentic misuse, AI-provider dependency, and AI data exposure. It produces an overall tier (Partial, Risk Informed, Repeatable, or Adaptive), a scored-dimension radar, and tailored next steps anchored to NIST AI RMF, NIST AI 600-1, OWASP LLM Top 10, and the major operational-resilience frameworks.

Question 2

How long does the AI-Threat Readiness Assessment take?

Accepted Answer

The short form is eight questions plus one sector-specific question and takes about six minutes. The long form is twenty-two questions plus one sector-specific question and takes about twelve minutes.

Question 3

Is the AI-Threat Readiness Assessment free?

Accepted Answer

Yes. The assessment is free and requires no login. The full per-dimension recommendations are unlocked by providing a work email so Bollwerk can send a recap of the result.

Question 4

Which frameworks does the AI-Threat Readiness Assessment map to?

Accepted Answer

Universal anchors include NIST AI RMF, NIST AI 600-1 (GenAI Profile), OWASP LLM Top 10, MITRE ATLAS, FATF digital-identity guidance, NIST SP 800-61, and ISO/IEC 42001. Jurisdiction-specific overlays add FINMA Guidance 08/2024 and 05/2025 (Switzerland), the EU AI Act and DORA (European Union), NCSC AI guidance (United Kingdom), and FFIEC and HHS / FDA references (United States).

Question 5

Which dimensions of AI threat exposure does the assessment cover?

Accepted Answer

Seven dimensions: payment-instruction integrity and manipulated business communication; synthetic-media and identity-attack readiness; AI-augmented phishing and social engineering; AI-driven fraud typologies in customer and transaction monitoring; security of own and vendor AI surfaces; AI incident response and operational resilience; and AI inventory, data exposure, and shadow-AI governance.

Question 6

Does Bollwerk store the assessment answers or use them to train AI models?

Accepted Answer

Answers are processed to produce the result. If the full report is unlocked, Bollwerk sends a recap to the submitter's email and a lead notification to its sales team. Assessment submissions are not used to train AI models.

AI-Threat Readiness Assessment

What it covers

Short form and long form

Jurisdiction overlays

Related