167-Signal AI Visibility Engine

LEGAL · AI-FIRST
EU · GDPR · AI Act
Compliance

EU AI Act Transparency

Last updated: 25 August 2026 · Operator: AIVENTURE S.R.L., Romania
On this page
  1. Summary
  2. Where AI is used
  3. Transparency — Article 50
  4. Risk classification
  5. Human oversight
  6. Data governance
  7. Accuracy and limitations
  8. Timeline
  9. Contact

This page sets out how 3webobs.com positions itself in relation to Regulation (EU) 2024/1689 — the European Artificial Intelligence Act — and what we disclose about the AI components of the Service.

1. Summary

The Service uses AI for one narrowly defined purpose: writing a short narrative summary of an already-computed technical result. Scores themselves are produced by deterministic rules applied to publicly observable evidence, not generated by a language model.

2. Where AI is used, and where it is not

ComponentAI involved?How it works
Fetching the website and its signal filesNoStandard HTTP requests
Detecting structured data, schema, crawler directivesNoDeterministic parsing
Scoring each signalNoFixed rules against observed evidence; unverifiable signals marked not applicable
Action plan orderingNoSorted by rule weight of failed signals
Written executive synthesisYesA general-purpose language model turns the computed result into a short narrative

2.1 Why the separation matters

2.1.1 Reproducibility

The same website in the same state produces the same result. A score is not a sample from a probability distribution.

2.1.2 No invented values

No score is generated by a model. Where evidence is unavailable, the signal is reported as not applicable rather than guessed.

3. Transparency — Article 50

3.1 The obligation

Article 50 of the AI Act requires that people be informed when they are interacting with AI-generated content.

3.2 Our disclosure

4. Risk classification

4.1 Prohibited practices — Article 5

Our assessment is that the Service does not fall within the prohibited practices of Article 5. It performs no social scoring, no subliminal manipulation, and no exploitation of vulnerability.

4.2 High-risk systems — Annex III

4.2.1 What we do not do

The Service does not perform biometric identification, emotion recognition, or evaluation of natural persons for access to education, employment, credit, essential services or law enforcement purposes.

4.2.2 What we do

The Service evaluates the technical configuration of a website. It does not profile individuals and does not produce decisions with legal or similarly significant effects on a person.

4.3 Our role

Where a language model is used for the narrative summary, the Service acts as a deployer of a general-purpose AI model supplied by a third party, and the transparency obligations above apply.

5. Human oversight

5.1 Reports are not decisions

Reports are diagnostic outputs. Every recommendation is intended to be reviewed and acted upon by a human.

5.2 No compliance verdict

We do not present the report as a compliance verdict, a certification, or a substitute for professional advice.

6. Data governance

6.1 Inputs

Inputs are publicly accessible website resources.

6.2 Training

6.2.1 No personal data used for training

No personal data is deliberately used to train any model.

6.2.2 Synthesis content

Content sent to the AI provider for synthesis is limited to the computed report; it is not used by us to build training datasets.

6.3 Handling of unverifiable signals

Where a signal cannot be verified, it is reported as not applicable rather than assigned a fabricated value.

7. Accuracy and limitations

7.1 What can distort a report

The report reflects what was publicly observable at the moment of the scan. It may be affected by:

7.2 What a report is not

Nothing in a report should be read as an assertion that a third party is or is not compliant with any legal obligation. A report identifies observable technical signals; it does not deliver a legal conclusion.

8. Timeline

8.1 Entry into force

The AI Act entered into force on 1 August 2024 and applies in stages.

8.2 Application stages

9. Contact and complaints

Questions about the AI components of this Service, or requests for explanation of a specific result: contact@5thelement.ai.

9.1 Note on legal status

This page states our own good-faith assessment of how the Service relates to the AI Act. It is not legal advice, and it is not a certification issued by any authority. If your own use case depends on a formal classification, obtain independent legal analysis.

Frequently Asked Questions

Ten questions that come up most often, answered plainly.

It measures how a website presents itself to AI systems, answer engines and search crawlers. The engine fetches the page, robots.txt, sitemap.xml and declared machine-readable signal files, then evaluates 167 signals across five dimensions: AI Signals, AEO, GEO, AIO and SEO.

No. Every score is produced by deterministic rules applied to evidence actually observed on the website. A language model is used only to write the short narrative summary at the end. This separation means the same website in the same state always produces the same score.

Because a signal that cannot be verified from public sources has not failed — it simply could not be tested. Marking it as failed would invent a verdict. Signals requiring paid external data sources or owner-level access are reported as not applicable, and the report states how many were tested.

SEO is the technical and content foundation used by search engines. AEO structures content so assistants extract direct answers. GEO makes the brand entity discoverable and citable by generative systems. AIO covers how AI systems understand, rank and reproduce content across the full stack.

No. The audit inspects only publicly served resources. Nothing is installed, no tracking code is added, no authentication is attempted, and no private or password-protected area is accessed.

The first audit is free. Subsequent audits are charged at the price displayed on the site at the time of ordering. A monthly plan adds score tracking, regression alerts and history. Machine-to-machine access for AI agents is priced separately under a written arrangement.

Yes. The engine exposes a POST endpoint at /audit that accepts a URL and returns a structured JSON report, plus a GET endpoint at /stats for service metadata. Negotiated machine-to-machine access, including rate limits, is agreed in writing.

No. Third-party AI systems change their behaviour without notice and no provider guarantees citation. The audit identifies observable technical signals that are within your control; it does not promise a ranking, a citation or a commercial outcome.

No. The report is a technical observation, not a certification and not a legal opinion. It identifies signals that justify further evaluation. For a formal classification under Regulation (EU) 2024/1689, obtain independent legal analysis.

AIVENTURE S.R.L., a company registered in Bucharest, Romania, European Union, operating under the 5thElement.ai brand. Registration details, contact routes and the full legal documentation are published on the site.

Quick links

Contact

Legal entity

© 2026 AIVENTURE S.R.L. · Audit·AI™ is a trademark of AIVENTURE S.R.L.
Reports are technical observations, not certifications or legal opinions.