The Classical Turing Test
In 1950, Alan Turing proposed a simple experiment: if a machine can converse so convincingly that a human judge cannot tell it apart from a real person, the machine can be said to “think.” This test shaped decades of AI research — but it was never designed to evaluate identity.
The Turing Test asks one question: can this machine be mistaken for a human in conversation? It evaluates presentation — not identity, biography, continuity, or authority.
The test at a glance
Interrogator
A human judge who asks questions through a text-only terminal
Human
A real person responding through an identical terminal
Machine
An AI system responding through an identical terminal
The interrogator sends questions to both unseen participants through identical text terminals. Neither the human nor the machine is identified — the judge must determine which is which based solely on their responses.
If the interrogator cannot reliably tell the machine apart from the human, the machine is said to pass the Turing Test.
Convincing conversation — nothing more.
A machine that passes the Turing Test has demonstrated that it can produce responses indistinguishable from a human's in a conversational setting. This is a meaningful benchmark for natural-language generation. But it tells us nothing about who the system is, where its knowledge comes from, whether it has a coherent history, or whether it will remain the same entity tomorrow.
Identity requires harder questions.
The Turing Test was not designed for the problems that arise when AI systems persist over time, represent specific people, migrate across models, or act with authority. These gaps motivate PCI's three identity tests.
No specific person
The test checks whether a machine seems generically human — not whether it faithfully represents one particular person.
No biography
It does not ask whether the machine has a coherent life story, credible memories, or appropriate ignorance.
No continuity
It says nothing about whether the system is the same identity after a model upgrade, migration, or branch.
No authority or governance
Passing the test does not establish who controls the system, what permissions it holds, or how its identity is governed.
PCI proposes three tests that ask harder questions.
Each test targets a gap the Turing Test was never designed to fill — referential fidelity to a specific person, biographical coherence, and governed continuity after change.
Human-Referential Fidelity Test
Does this system match this particular person?
Situated Identity Test
Is this agent's behavior uniquely attributable to its actual developmental lineage?
Cognitive Continuity Test
Is this still the same governed identity after change?