Legal
Terms
Status of this website
This website describes a proposed initiative. Andorra AI Safe Haven is in design and stakeholder consultation. Nothing here constitutes an offer, contract, or a commitment to deliver any specific capability, timeline or service level.
No endorsement claimed
No government, bank, regulator, standards body or certification authority endorses, certifies or funds this initiative. Government endorsement, regulatory recognition and certification are objectives to explore.
Architecture and capability statements
Descriptions of the Temporal Replay Engine, AI Evidence Envelope, cryptographic controls, assurance levels and sovereign deployment are architecture targets for a proposed platform. They are not statements of deployed, validated or certified capability. Exact computational reproducibility of AI events may depend on the continued availability and deterministic behaviour of external models and services.
Source material
References to bitemporal data management, Gartner research from 2008 and the historic Temporal Replay Engine concept demonstrated by Oracle Consulting UK are cited as background. The paper concerned explicitly states that TRE was not an Oracle product.