# AI PSYOPS taxonomy methodology

This taxonomy synthesizes twelve supplied interdisciplinary reports. It preserves their distinctions among historical precedent, confirmed real-world use, demonstrated capability, emerging risk, contested evidence, and speculation. The release did not independently reverify time-sensitive external claims.

## Principles

- Engagement, impressions, reach, virality, or account count are not treated as proof of persuasion or behavior change.
- Attribution is presented as confirmed, alleged, inferred, contested, or unknown according to the reports.
- Fluent output does not establish strategic competence, consciousness, or autonomy.
- Detection indicators are suggestive unless the underlying evidence supports a stronger conclusion.
- Case studies demonstrate possibility or documented use; they do not automatically establish prevalence.
- Legal interpretations are summarized from the reports and may vary by jurisdiction or change over time.
- Operational details are abstracted into capability, risk, detection, and governance concepts.

## Evidence legend

- **Documented current use** — The report identifies public evidence that the capability was used in a real setting.
- **Demonstrated capability** — The capability was shown in a controlled study, prototype, or bounded deployment.
- **Emerging capability** — Components exist, but reliable or broad real-world use remains limited.
- **Primarily prospective** — The report treats the capability mainly as a future or conceptual risk.
- **Evidence contested or incomplete** — Evidence, effect size, attribution, or interpretation is disputed or methodologically limited.
- **Mixed evidence** — Different parts of the category have different maturity or evidentiary support.

## Safety boundary

The public section excludes operational targeting, scripts, vulnerability scoring, deployment instructions, evasion, and campaign optimization. It is intended for research, defense, governance, and public education.

## Editorial-verification additions

All 91 selected references have resolution records. Every substantive category claim is linked in the claim-to-source matrix. Ten evidence tensions remain public and require human adjudication. This automation does not constitute human editorial or publication acceptance.

## Citation integrity

The v100.0.45 release assigns a stable identifier to every public claim, renders claim links beside the corresponding category content, and publishes a deterministic graph linking 12 categories, 501 claims, and 91 selected sources. The graph checks internal traceability only; it does not establish source truth, currentness, legal accuracy, causal effect, human review, or publication acceptance.