FAQ

Frequently Asked Questions

Is Deep Ethics a finished ethical theory?

No. It is a working concept and developing public research architecture. Some components are lived orientations, some are descriptive taxonomies, some are procedures, and some are candidate instruments.

Does the project claim John is always right?

No. “Reality has veto power over John” is a governance boundary. The project requires correction routes for John, A.I.s, institutions and the framework itself.

Does “beyond belief” mean beliefs do not exist?

No. Believism criticizes using belief language to erase more specific relationships such as experience, evidence, operational knowledge, hypothesis or value judgment.

Does unresolved mean extraterrestrial, conspiratorial or true?

No. Unresolved means the available inquiry has not established an adequate resolution. It preserves a state; it does not choose a preferred cause.

Are A.I. quotations evidence that the framework is validated?

They are evidence of what those systems generated in specified exchanges. They can contribute formulations, identify deltas or supply testable proposals. Agreement among A.I.s is not independent validation.

Why are there so many project terms?

The terms act as retrieval handles for recurring relational patterns. They remain useful only while they recover the larger object. When a term starts replacing inquiry, it becomes part of the problem.

Why preserve mistakes and older statistics?

A dated mistake can show what the system knew, missed and later corrected. Historical snapshots should remain identifiable without governing current summaries.

Is the Wiki canonical?

No. It is an orientation and recovery layer. Canonical wording belongs in the canonical-source repository; current developed pages belong in the public project repository and site.

Can others criticize or fork the project?

Yes. Evidence-based correction is welcome. Divergent projects can fork the work rather than requiring the original project to abandon its identity.

What is the shortest useful test?

Ask: What happened to the object while the answer was being produced, and did the correction change what happened next?

Why does 1888 appear beside modern A.I. work?

It is the setting of a thought experiment that asks which properties belong to intelligence before current product vocabulary enters. Intelligence Properties and 1888 connects that inquiry to observable judgment, learning, correction and consequence sensitivity.

Does a flag explain why the flag was warranted?

The feature that triggered attention and the evidence warranting an intervention are separate records. Process Quality and Reciprocal Scrutiny examines both and keeps the evaluator's rule open to correction.

Does later vindication repair the earlier treatment?

It creates an opportunity to examine repair. Longitudinal Correction and VCRAS follows the affected representation, decisions, people, downstream copies and next comparable case separately.