The Process
Three steps to full compliance visibility
Connect your documentation sources
Upload files, paste a URL or connect your cloud in seconds.
GuardianAI accepts PDFs, DOCX and text files regardless of size. You can also paste the URL of a web policy, legal notice or terms page and the system extracts the content automatically. If your documentation lives in the cloud, connect Google Drive, OneDrive, SharePoint or Dropbox via OAuth in one click — no need to download or re-upload files. For teams working in Slack, the Slack connector scans messages and shared files in the channels you configure. The entire ingestion process is encrypted with TLS 1.3 and AES-256; your documents are never used to train models.
- PDF, DOCX, TXT — no page limit
- Direct URL to any public web policy
- Google Drive, OneDrive, SharePoint, Dropbox via OAuth
- Slack channels with granular channel selection
- TLS 1.3 + AES-256 encryption from the first byte
AI analysis article by article
The engine analyzes every paragraph against EU AI Act, GDPR & ISO 42001 in under 2 minutes.
Once the document is loaded, GuardianAI's engine segments it and analyzes it paragraph by paragraph against the EU AI Act, GDPR and ISO 42001. For each relevant article, the system determines whether sufficient evidence, partial evidence or a total absence of compliance exists. Every finding is classified by severity (Critical, High, Medium, Low) and anchored to the exact textual quote from your document that creates the gap — no generalities, no opinions. The full analysis takes under 2 minutes for documents up to 200 pages.
- Coverage of all applicable EU AI Act articles
- Simultaneous analysis against GDPR & ISO 42001
- Severity classification: Critical / High / Medium / Low
- Exact textual quote of the paragraph creating each gap
- Under 2 minutes for documents up to 200 pages
Act with a clear plan and automate monitoring
Prioritized report, AI chat to resolve questions, and continuous automatic patrolling.
The analysis result is an interactive dashboard where every gap has an action priority, a suggested fix and the exact regulatory article reference. You can ask the AI assistant any question about findings or the regulation — it replies with context from your own document. Once you've fixed the gaps, generate an executive report as PDF or CSV ready to share with your legal team, investors or audit team. To stop relying on manual scans, configure active patrolling: GuardianAI automatically rescans your sources whenever it detects changes and alerts you in Slack or by email if a new risk appears.
- Dashboard with gaps ordered by action priority
- Suggested fix for each finding with article reference
- AI chat with context from your documentation and the regulation
- PDF/CSV export ready for auditors and investors
- Active patrolling: automatic alerts on changes or new risks
Capabilities
Everything GuardianAI can do for you
Document and URL scanning
GuardianAI's ingestion engine supports any standard document format — PDF, DOCX, TXT — and extracts text while preserving the original section and paragraph structure. For URLs, a lightweight crawler visits the page, removes navigation noise and extracts only the relevant content. This allows analyzing privacy policies, terms of service or AI system notices published on the web without any additional manual steps.
- PDF up to 200 pages in < 2 min
- Semantic extraction of web URLs
- Automatic language detection
- No loss of section structure
Compliance score 0–100
Each scan produces a numeric score from 0 to 100 reflecting the percentage of covered requirements weighted by regulatory importance. A score above 80 indicates a solid compliance posture; between 60 and 80 there are gaps requiring attention; below 60 there is elevated non-compliance risk. The score evolves with each new scan, allowing you to track progress over time in the Analytics tab.
- 0–100 weighted by regulatory relevance
- Visual threshold: green / amber / red
- History of evolution per document
- Comparison between document versions
Automatic risk classification
The EU AI Act classifies AI systems into four levels: Prohibited, High Risk, Limited Risk and Minimal Risk. GuardianAI automatically applies the Annex III and Article 6 criteria to determine which category your system falls into based on submitted documentation. This classification determines which obligations apply: from EU database registration to human oversight, risk management and transparency requirements.
- Automatic application of Annex III criteria
- Classification: Prohibited / High / Limited / Minimal
- Obligation map per risk level
- Reference to the exact article defining the category
Findings with exact evidence
Every identified gap comes with three pieces of information: the exact textual quote from your document creating the problem, the article and paragraph of the regulation being breached, and a concrete actionable fix. No generalities. If Art. 13 requires the system to notify the user they are interacting with AI and your document doesn't mention it, the finding tells you exactly which section to add and how to draft it.
- Textual quote from the document creating the gap
- Exact reference: Article + paragraph + regulation
- Drafted fix ready to implement
- Severity: Critical / High / Medium / Low
Native connectors: Drive, Slack, Dropbox & more
GuardianAI connects directly with the tools your team already uses via secure OAuth. Google Drive and OneDrive allow selecting specific folders for scanning. SharePoint facilitates analysis of corporate documentation without moving files. Dropbox automatically syncs new files. The Slack connector monitors channels where policies are shared or system changes are discussed. New finding notifications are sent directly to the Slack channel you choose.
- Google Drive and OneDrive with folder selection
- SharePoint for corporate documentation
- Dropbox with automatic sync
- Slack: channel scanning + alert notifications
Active patrolling 24/7
Active patrolling turns GuardianAI into a continuous monitoring system rather than a point-in-time audit tool. Configure which folders, URLs or channels you want to watch and how often (daily, weekly or in real time for detected changes). When GuardianAI detects a modification in documentation or a new file, it automatically launches a new analysis and only notifies you if a new risk appears or an existing one worsens. This eliminates the need for periodic manual reviews.
- Configurable frequency: daily, weekly or real-time
- Alert only when new or worsened risk appears
- Coverage of multiple sources simultaneously
- Auditable patrol history
AI assistant with regulatory context
The built-in AI chat combines comprehensive knowledge of the EU AI Act, GDPR and ISO 42001 with the specific content of your scanned documents. You can ask it directly what Art. 17 means for your specific case, how to draft a human oversight policy, or what evidence is missing to resolve a critical finding. Answers are precise, contextualized and always cite the regulatory source. It's not a generic chatbot: it understands your specific situation.
- Combined context: your doc + full regulatory text
- Answers citing article and paragraph
- Drafting suggestions to cover gaps
- Unlimited questions on Professional+ plans
Exportable reports for auditors and investors
GuardianAI generates professional PDF reports and structured CSV data ready for three distinct audiences: internal compliance teams that need technical detail, investors or due diligence requiring an executive view of the regulatory status, and external auditors needing traceable evidence. The PDF report includes executive summary, score, gap list by severity, and action plan with regulatory references. The CSV facilitates integration with spreadsheets or risk management tools.
- Executive PDF with summary, score and action plan
- Structured CSV for integration with other tools
- Full traceability: date, version, source analyzed
- Ready for due diligence, Series A/B and external audits
Who is it for?
Built for the teams that make compliance happen
Compliance & Legal Teams
Reduce weeks of manual review to minutes. Get findings with exact evidence you can present directly to the CISO or Board without additional interpretation work.
CTOs & Product Teams
Integrate compliance analysis into the development cycle. Scan technical documentation before each launch and detect gaps before they reach an audit.
DPOs & Regulatory Consultancies
Manage multiple clients from a centralized dashboard (Business+ plan). Generate branded reports for each client and configure independent patrolling per organization.
Pre-investment or Pre-launch Startups
Demonstrate solid regulatory posture in Series A/B due diligence. Detect the critical EU AI Act gaps before an investor or external auditor does.
FAQ