EU AI Act
Articles 14 and 26
Human oversight records for the AI systems you operate
What the regulation requires
Article 14 sets human-oversight design requirements for high-risk AI systems. Article 26 asks deployers of those systems to assign oversight to people with the necessary competence, training and authority, and to use the system according to its instructions. Which obligations apply to your system is a legal assessment for your organisation.
How Siglato supports it
Assigned oversight
Name the people responsible for each registered system, with their authority, training evidence and escalation path.
Review queue
Findings that need human judgement go to a review queue. Each acknowledgement, decision, rationale, override or suspension recommendation is recorded. Siglato records oversight; it does not stop or control your AI system.
Activity record
Events your application sends through the HTTP API or Python SDK are stored in a tamper-evident audit trail and linked to the reviews they triggered.
Evidence report
Export JSON, HTML or PDF evidence with its limits: what Siglato received, known gaps, and periods for which completeness cannot be established.
What the evidence looks like
Note
Integrity verified on your infrastructure. The report lists received records and known gaps.
The records shown are illustrations. Siglato is in early access; documentation is delivered with the evaluation bundle. Request access.