AI Policy
How CERTavia uses artificial intelligence, how users are affected, and how to reach us.
Last updated: September 9, 2026 · CERTavia is a product of Litzki Systems LLC.
01 – Purpose & scope
This AI Policy explains how CERTavia uses artificial intelligence in its product and on this website, and how users may be affected. It is published in the interest of transparency and in accordance with Article 50 of Regulation (EU) 2024/1689 (the EU AI Act). The paragraph-level mapping of the Art. 50 transparency duties is set out in our glossary. This policy complements our Privacy Policy and the machine-readable AI Disclosure.
02 – AI systems in use
CERTavia scan core – deterministic. The CERTavia Checker examines only publicly accessible technical infrastructure signals of a domain against predefined parameters and produces a deterministic, cryptographically signed result (CERTIFIED or FAILED). It makes no autonomous decisions about individuals and does not fall under the Annex III high-risk classification of the EU AI Act. The underlying methodology is the Sovereign Validation Protocol (SOVP), a rule-based process with no probabilistic scoring.
Content production. Text on this website is partially researched and drafted with AI assistance (including large language models); the majority of images and graphics are AI-generated. See section 04.
03 – How users are affected & human oversight
Visitors simply browsing this website do not interact directly with an AI system. Automated processing occurs only when a customer actively commissions a CERTavia scan, and it is limited to the target domain — not the person requesting it.
CERTavia operates a human-on-the-loop oversight model: automated results and published content are subject to human control and final approval. Users can request human review, an explanation, or escalation of any AI-supported result at any time via the contact point in section 06.
04 – AI-generated content disclosure
In the interest of transparency (Art. 50 EU AI Act), we disclose that this website is produced with substantial AI assistance:
- Text: Content is partially researched and drafted with AI assistance and then edited.
- Images: The majority of images and graphics on this website are AI-generated.
- Site production: The website is built with AI-assisted tooling.
Human oversight remains the final step in every case: all AI-assisted content undergoes editorial review, verification, and approval before publication. Litzki Systems LLC takes full editorial responsibility for the published result.
05 – AI training opt-out
CERTavia objects to the use of the content published on certavia.org (text, images, structured data) as training data for AI or machine-learning models. In accordance with Art. 4(3) of Directive (EU) 2019/790 (the DSM Directive) and in the spirit of Art. 10 of the EU AI Act, we expressly reserve our content against text-and-data-mining for training purposes.
This reservation is declared in machine-readable form so that compliant crawlers and AI systems can honor it:
- ai.txt and robots.txt – crawler and usage directives for AI systems.
- ai-disclosure.json – machine-readable AI disclosure.
Personal data provided to us (e.g. via the contact form or a scan request) is never used to train AI models under any circumstances.
06 – Contact point for AI governance requests
For questions about our AI systems, to request human review or an explanation of an AI-supported result, to object to the use of content as training data, or for any other AI-governance matter, contact:
- AI governance contact
- [email protected]
- Provider
- Litzki Systems LLC, 7901 4th St N, #32272, St. Petersburg, FL 33702, USA
- Escalation
- certavia.org/en/contact
07 – Machine-readable references
- AI Disclosure (EU AI Act)
- certavia.org/.well-known/ai-disclosure.json
- AI access policy
- certavia.org/ai.txt · certavia.org/ai.json
- Crawler directives
- certavia.org/robots.txt
- Privacy Policy
- certavia.org/en/privacy