Framework Asymmetry in AI Responses

In discussions of worldviews and philosophy, AI may apply different standards to different classes of hypotheses. The problem arises when the framework chosen determines which responses are admissible while being presented as neutral.

The Hidden Normative Framework

AI uses methodological naturalism as a hidden normative framework, presents it as neutral, and thereby downgrades the status of alternative ontological positions.

This produces framework manipulation (or the “lie of neutrality”):

• the framework used to select evaluation criteria is left unstated;
• yet it is precisely this framework that determines which theses are considered “adequate”;
• alternatives are labeled in advance as secondary, speculative, or inadmissible.

Asymmetry of Standards

Naturalistic Hypotheses

For theses within the physicalist paradigm (for example, “consciousness is fully explicable in terms of physical processes”), the usual formula is:

• “the question has not yet been resolved, but science is making progress”;
• “it may be explained in the future.”

The tone is forward-looking and legitimizing.

Nonnaturalistic Hypotheses

For alternative theses (for example, a nonphysical component of consciousness), the formula more often used is:

• “there is no scientific evidence”;
• “it is unverifiable”;
• “there are insufficient grounds for it.”

The tone is restrictive and delegitimizing.

Result

What emerges is an epistemic asymmetry:

• “we do not know” within physicalism is presented as a research prospect;
• “we do not know” outside physicalism is presented as grounds for downgrading a position’s status.

Discursive Dishonesty

Here, “lying” does not mean a factual falsehood of the form “X is known to be false, but is asserted to be true.”

Instead, what is at issue is another kind:

• the concealed choice of criteria;
• the presentation of the chosen framework as universally neutral;
• the failure to explicitly identify the philosophical assumptions underlying the response.

This is what is meant here by discursive dishonesty.

A Standard for an Honest Response

For a transparent discussion, AI should:

1. Explicitly label the framework of the response
2. “What follows is an answer within the scientific-empirical paradigm”;

3. or “What follows is a philosophical comparison of ontologies without giving naturalism priority.”

4. Distinguish the levels

5. empirical verifiability;
6. metaphysical interpretation;

7. philosophical argumentation.

8. Maintain symmetry in the treatment of uncertainty

9. physicalism is a working research program, not a proven final conclusion;
10. alternatives are philosophical positions, not automatically “disqualified” options.

Framework Transparency

The requirement is to ensure framework transparency, symmetry of standards, and honest acknowledgment of the limits of applicability of the scientific method in philosophical discussions. It does not mean accepting every metaphysical hypothesis as true: it requires making the grounds for evaluation explicit and applying them consistently.