EU AI Act AssistantThe GRC expert your AI calls.

Free AI guidance for EU AI Act compliance. No account needed.

Ask here, or call it from Claude, Codex or Cursor.

Supplier agreement reviewSampleCloud hosting contract.pdf
3 findings, all decided
FindingRequirementDecisionWho
HighThe supplier does not have to tell you about security incidents.Suggested: notice within 24 hours of becoming aware, and a duty to cooperate.A.5.20Suggestion acceptedDecided by youYou
MediumSubcontractors are not bound by the same security terms.Suggested: flow the security clauses down to every subcontractor.A.5.21Accepted with your wordingDecided by youYou
LowYour right to audit is limited to their annual SOC 2 report.Suggested: add a right to audit after a major incident.A.5.22RejectedRisk accepted for this supplier. Revisit at renewal.Decided by youYou

Sample review. Reviews and the record of each decision are part of a paid plan.

About the EU AI Act assistant

Ask about EU AI Act concepts: the risk-based classification (prohibited practices, high-risk systems, transparency risk, and minimal risk), provider and deployer roles, the high-risk requirements such as risk management, data governance, technical documentation, and human oversight, transparency rules for AI-generated content, and the obligations for general-purpose AI models.

  • Work through which risk category your AI system falls into and whether the EU AI Act applies to you as a provider or a deployer.
  • Plan the high-risk requirements: risk management, data governance, technical documentation, logging, human oversight, and conformity assessment.
  • Understand the transparency rules for AI-generated content and chatbots, the general-purpose AI model obligations, and how the Act relates to ISO 42001.

It gives implementation guidance to speed up your work. It is not legal advice, and the Act's obligations phase in over time, so confirm the current dates and specifics with a qualified professional.

Read the full EU AI Act guide

Example questions you can ask

  • We fine-tune and release a general-purpose AI model. Outline the Article 53 copyright policy and the public summary of training content we need.
  • Draft an AI literacy plan under Article 4 of the EU AI Act for our sales and support teams who use AI tools every day.
  • Map the Article 17 quality management system points (a) to (m) to the ISO/IEC 42001 clauses we already have in place.

Frequently asked questions

Can it help me document why an Annex III system is not high-risk under Article 6(3)?

Yes. It tests your system against the four Article 6(3) conditions: a narrow procedural task, improving the result of a completed human activity, detecting decision patterns without replacing human assessment absent proper review, or a preparatory task. It flags the hard stop that a system profiling natural persons stays high-risk. It then drafts the assessment record Article 6(4) expects before you place the system on the market, and reminds you that it still needs registering under Article 49(2).

Can it draft a fundamental rights impact assessment under Article 27?

If you are a public body or another deployer named in Article 27(1), it drafts the structure and much of the content in the chat: how and how often you will use the system, affected groups of people, specific risks of harm drawing on the provider's Article 13 information, your human oversight measures, and your response if risks materialise. Article 27(4) lets you cross-reference a GDPR Article 35 DPIA that already covers a point. Describe affected groups in general terms, not with real personal data.

Does it reflect the Digital Omnibus changes to the AI Act?

Yes. Its notes follow Regulation (EU) 2026/1744, the Digital Omnibus on AI, in force since 27 July 2026, and explain what it changed: revised Article 113 dates for the high-risk chapter, new Article 5 bans from 2 December 2026 on AI producing realistic intimate depictions of identifiable people without consent or child sexual abuse material, bias-correction processing of special-category data moved from Article 10(5) to a new Article 4a, and a simplified technical documentation form for SMEs, start-ups and small mid-caps.

Can it help with the Article 50 transparency rules for our chatbot and generated content?

Article 50 has applied since 2 August 2026. It helps you word the notice telling users they are dealing with an AI system, unless obvious, given at the latest at first interaction (Article 50(1) and (5)). For synthetic audio, images, video or text, it explains the machine-readable marking duty in Article 50(2) and the extra time in Article 111(4): for systems placed on the market before 2 August 2026 the deadline is 2 December 2026. Deepfake disclosure under Article 50(4) falls on deployers.

Can it help us set up serious incident reporting for a high-risk AI system?

It can draft the procedure a provider needs under Article 73: report to the market surveillance authority where the incident occurred immediately once a causal link is established, and at the latest 15 days after becoming aware; within 2 days for a widespread infringement or a serious and irreversible disruption of critical infrastructure; and within 10 days where a person dies. It ties this to your Article 72 post-market monitoring plan and to the investigation and corrective action that Article 73(6) expects.