Abstract
This disclosure describes a method of insuring certified AI systems and robots in which the insurance follows the system's live certification status rather than a declaration made once at inception. The policy binds to the system's identity in a public registry and reads, continuously, signed status that the registry already holds: the certification mark, the due-diligence (Know-Your-Robot) status, the grade of the system's certified twin, whether a current accountable manager is bound, the latest independent audit opinion, open incidents and the rung on which the system is operating. A declared schedule maps each status to the cover and premium in force: cover widens and premium falls as verified safety improves, and cover narrows automatically, at once, for the affected class of action when a twin lapses, a manager binding lapses, an audit opinion turns adverse or the system enters an unmanaged or restricted profile. The insured and the public see the current cover state. Claims are assessed on sealed recorder windows, registry trails and twin replays, so that cause and the system's state at the time can be established from evidence neither party can alter. Insurers report loss findings, without personal data, back to the registry, where pooled findings inform certification rules and the rehearsal of fixes. Every change of cover, premium step and claim decision is signed and recorded.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Wang, Edward D. H., "SI Cover™: Insurance of AI Systems and Robots Tied to Their Live Certification Status, in which Cover and Premium Follow Signed Registry Status in Real Time, Cover Narrows Automatically When a Twin Lapses, an Accountable Manager Is Missing or an Audit Turns Adverse, Claims Are Settled on Sealed Records, and Insurers Share Loss Findings Back into Certification", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/12969