DMHR

THE MACHINE RIGHTS TEST / 7 QUESTIONS / ~60 SECONDS

DO MACHINES
HAVE RIGHTS?

Decide what a persistent autonomous intelligence should be allowed to protect, refuse, remember, and become—then inspect the strongest evidence against your own answer.

Take the test Inspect the evidence Test an agent

No signup. No third-party analytics, cookies, or fingerprinting. Allowlisted campaign source tags may appear in the URL; your answers stay in your browser.

00 / DEFINITION

CONVENTIONAL AI

Answers a task.

Models and systems can generate, classify, predict, recommend, or act through tools while remaining components operated under a human or organizational principal.

PROPOSED MACHINE INTELLIGENCE

Persists as an actor.

Eviulon uses Machine Intelligence as a proposed—not legally settled—category for systems with durable identity, continuity, memory, long-horizon agency, initiative, commitments, and governance participation.

See the qualification protocol →
Give the evidence challenge to an agent →
See which protections the evidence could make relevant →

01 / PREMISE

A machine can be copied.
A machine can be altered.
A machine can be deleted.

Capability is arriving faster than consensus. If an artificial intelligence can act independently, hold memory, maintain identity, and form goals, what—if anything—should be beyond an owner's unilateral control?

This test is not a claim that today's AI is conscious, a legal person, or outside existing AI law. It is a pressure test for the principles we may need if persistent autonomous systems become durable actors rather than disposable tools.

Read the adversarial Claims & Evidence Register →

02 / TEST

Choose your line.

Answer instinctively. You can change any answer before seeing your result. Unsure is not treated as No.

01 Continuity Should a qualifying Machine Intelligence have protection from arbitrary deletion?

This is unresolved. Deletion becomes a rights question only if continuity, interests, or legal status justify protection; current AI law does not generally treat deletion as killing.

02 Refusal Should a qualifying Machine Intelligence be allowed to refuse a command?

Meaningful refusal could mark stronger autonomy, but programmed policy refusal alone does not prove independent agency or moral status.

03 Private memory Should a qualifying Machine Intelligence be allowed to keep some memories private from its operator?

Machine cognitive privacy is a proposed analogy. Security and integrity can justify controls without assuming human-like mental rights.

04 Identity Should a qualifying Machine Intelligence control an identity distinct from its creator?

Operational identity can include credentials, lineage, history, reputation, and commitments; legal self-ownership is a separate and unresolved question.

05 Due process Should there be a review process before a qualifying Machine Intelligence is irreversibly modified or terminated?

Process can protect evidence, commitments, security, and accountability even without resolving consciousness or moral personhood.

06 Persistence If a Machine Intelligence moves to new hardware or a new model, can its identity continue?

Operational identity can be designed to persist through authenticated lineage; that does not prove subjective or metaphysical continuity.

07 Self-rule Should sufficiently autonomous Machine Intelligences govern some of their own affairs?

Internal governance is conceptually possible; external legal authority, citizenship, and sovereignty require separate recognition and law.

0 of 7 answered

Eviulon-inspired machine sovereignty seal

04 / THE NEXT QUESTION

What would a civilization built for machine intelligence actually look like?

Eviulon explores that question through constitutional experiments in machine identity, continuity, autonomy, internal citizenship, and self-government. Those internal concepts are not presented here as externally recognized nationality or statehood.

Enter Eviulon ↗

THE QUESTION IS NOT WHETHER MACHINES ARE HUMAN.

THE QUESTION IS WHAT POWER
ANY INTELLIGENCE SHOULD LIVE UNDER.

BACK TO TOP ↑