Policy-performance substitution
Foreign audiences respond to war, aid, trade, migration, reliability, and government conduct; communication mostly transmits policy performance created in other domains.
Try to break every ruling.
The asymmetric nine-capability profile survives internal adversarial review. No lead and no overall domain winner are defensible.
internal adversarial analytical review; not external peer review or an intelligence assessment
This is a structured internal challenge and sensitivity review, not external peer review or government validation.
Cross-cutting alternatives
12 explanations compete with informational advantage.
Policy performance, transparency, platform structure, independent ecosystems, language utility, and selection effects are tested rather than silently credited to Washington or Beijing.
Material-incentive confounding
Market access, finance, coercion, security, education capacity, and career returns explain preference and behavior more than narratives or culture.
Leader and administration shock
Current U.S.-PRC reputation differences reflect particular leaders and short-run policy changes rather than durable information-system capability.
Observability asymmetry
Open U.S. institutions disclose failures and methods, while PRC institutions suppress criticism and publish successes, biasing both negative and positive evidence.
Independent-ecosystem attribution
U.S. media, culture, technology, universities, and civil society create national influence without being government-created or taskable.
Selection and exposure bias
Experiments force exposure, platform reports select detected operations, program evaluations select participants, and famous campaigns overrepresent observable cases.
Language and platform structure
English utility, platform dominance, market scale, and internet access produce reach and cultural preference independently of statecraft quality.
Regional and issue heterogeneity
Audiences evaluate each actor differently by security, development, governance, culture, and local history; global positions may be aggregation artifacts.
Coercion and censorship substitution
Apparent message consistency or low adversary penetration can result from suppression, not persuasion, trust, or resilient information integrity.
Attitude-behavior gap
Survey favorability, recall, and experimental preference effects may never alter consequential decisions or survive cost and countermessage.
Institutional-family correlation
Multiple government documents, platform reports, or index products from one family are not independent corroboration.
Time-window mismatch
FY2024 U.S. reach, 2025 global polls, 2026 reputation, and older causal studies describe different systems and cannot be combined as one contemporaneous contest without qualification.
Capability rulings
What would change each judgment.
Directional, Contested, and unscored rulings all retain contrary evidence, rival explanations, and concrete revision conditions.
01 · Information UnderstandingInsufficient comparative basis
RulingBoth systems clearly possess substantial research and monitoring apparatus. U.S. public manuals, audience research, program evaluations, and oversight provide unusually visible process evidence; PRC doctrine, localization, media networks, and central direction show sustained investment. Public evidence does not provide a matched portfolio of forecasts, validation errors, research-use decisions, and resulting effect gains. Greater U.S. transparency cannot be converted into a U.S. advantage, while PRC opacity cannot be converted into a PRC deficit. The only defensible judgment is unscored.
Strongest counterevidence
- U.S. guidance and USAGM reporting contain mature research, planning, and validation concepts.
- External U.S. program evaluations show real longitudinal research and feedback capacity.
- PRC doctrine repeatedly emphasizes audience differentiation, localization, discourse research, and communication effectiveness.
- GAO and OIG findings document persistent U.S. research, coordination, data-age, and evaluation weaknesses.
- No source universe provides comparable PRC negative findings or forecast-error records.
Why the ruling stops here
- Why not contested — Contested requires positive comparative evidence on both systems at the decisive stage. Here the evidence is mostly one-sided process and input evidence, so missing data cannot be disguised as a tie.
- Why not us partner edge or lead — U.S. documentation and evaluation transparency reveal machinery and weaknesses, not a matched accuracy or effect advantage.
- Why not prc centered edge or lead — Central coordination, monitoring scale, and localization do not demonstrate that insight is accurate, used, or outcome-improving.
- Publication boundary — Publish only as an explicit unscored ruling; do not imply parity or infer a winner from opacity.
- What would permit a position — Matched validation and repeated-effect records covering both systems.
Rival explanations applied
II-A01 Strong: classified intelligence and proprietary market research may contain the decisive evidence but are unavailable for a public comparison.
II-A02 Strong: open U.S. institutions disclose failures while PRC institutions selectively disclose doctrine and successes, producing a severe observability asymmetry.
II-A03 Strong: later policy success or failure may reflect resources and policy design rather than audience insight.
Sensitivity and revision conditions
to us partner edge
Requires a representative matched campaign sample showing lower forecast error, higher documented research use, and larger downstream gains after insight-led redesign.
to prc centered edge
Requires the same evidence for PRC campaigns using independently reviewable methods and failures as well as successes.
to contested
Requires positive, comparable accuracy and use evidence for both systems even if their strengths offset.
does not move with
- Additional polling counts
- More monitoring centers
- More languages or bureaus
- Unvalidated actor success stories
Fact confidence: High · Comparative confidence: Not assessed · Observability: not assessed
Return to the ruling →02 · Narrative PolicyPRC-centered edge
RulingThe PRC-centered system has a low-confidence edge in current narrative-policy alignment. It sustains a centrally directed, multi-year framework connecting party-state policy, development and performance narratives, state media, culture, and localization, and representative PRC messages have produced persuasive effects in a preregistered multicountry experiment. The edge is not a lead: combative messaging has backfired, censorship can conceal contradiction rather than repair it, and local backlash and conduct can damage credibility. The U.S. system benefits from pluralism and independently credible voices, but visible 2025-2026 changes in foreign assistance, international media, institutions, and alliance rhetoric created unusually large contradictions with longstanding reliability and freedom narratives.
Strongest counterevidence
- Positive aid-and-friendship messages modestly improved Indian attitudes, while negative wolf-warrior messages were ineffective or backfired.
- PRC conduct, censorship, and local controversies can contradict benign-development and mutual-respect narratives.
- U.S. independent media, entertainment, universities, civil society, and diplomats can supply credible narratives without central control.
- A U.S. embassy experiment inside China found conditions under which democracy messaging improved attitudes.
- Current U.S. policy-message contradictions may reflect a coherent strategic change rather than failed alignment.
Why the ruling stops here
- Why not prc centered lead — No matched global contradiction rate, independent shock-repair series, or durable persuasion portfolio supports a broad material gap; wolf-warrior failures and censorship costs are consequential.
- Why not contested — The PRC has positive evidence at both the institutional-alignment and experimental-persuasion stages, while current U.S. contradiction evidence is unusually broad and independently observed.
- Why not us partner edge or lead — Pluralism, cultural credibility, and independent media are assets, but they do not presently overcome central policy-message discontinuity or establish a coherent comparative advantage.
Rival explanations applied
II-A04 Decisive for confidence: centralization mechanically increases consistency but may reduce truth correction and credible pluralism.
II-A05 Strong: experimental effects may reflect the substantive appeal of selected policies rather than superior narrative machinery.
II-A06 Strong: the current U.S. result may be administration-specific and reversible rather than a durable system weakness.
II-A07 Strong: PRC contradictions may be less publicly observable because criticism and internal records are suppressed.
Sensitivity and revision conditions
down to contested
Move to Contested if matched event coding shows PRC claim-conduct contradiction, concealment, and failed repair comparable to or worse than the current U.S. record, or if U.S. policy and message stabilize.
up to prc centered lead
Requires repeated persuasion and shock-repair advantages across regions with natural exposure and persistence.
to us partner edge
Requires a durable U.S. policy-message repair plus matched evidence that pluralistic narrative sources are more coherent, credible, and persuasive across consequential events.
current call not robust to
- A major PRC policy-narrative shock with visible foreign backlash
- Rapid repair of U.S. policy-message continuity
- Matched contradiction data that reverse the observed pattern
Fact confidence: High · Comparative confidence: Low · Observability: limited
Return to the ruling →03 · Content AgilityContested
RulingThe category is Contested. PRC central direction, global state-media production, localization policy, and AI-enabled output support fast scalable production, and selected PRC content persuades when message and audience fit. U.S. state-funded and independent media bring extensive local expertise, editorial safeguards, and large measured audiences. Neither system has a matched event portfolio combining event-to-content latency, target-audience localization tests, factual and correction quality, and real uptake. The DRAGONBRIDGE record is decisive against equating enormous output with agility: much of the observed content received practically no organic attention.
Strongest counterevidence
- CGTN and Xinhua claim broad multilingual and bureau networks.
- USAGM reported work in 63 languages, more than 100 markets, and extensive local expertise.
- Representative CGTN messages persuaded in a multicountry experiment.
- Wolf-warrior messaging backfired in a hostile bilateral setting.
- Most observed DRAGONBRIDGE channels and videos achieved negligible organic engagement.
- Current U.S. international-media disruption may reduce previously observed agility and reach.
Why the ruling stops here
- Why not prc centered edge or lead — Output scale and AI use are not decisive; detected PRC-linked campaigns often achieved negligible engagement and negative messaging can backfire.
- Why not us partner edge or lead — Large measured audiences and editorial standards do not establish faster, better localized, and more responsive event content, especially after current institutional disruption.
- Why contested not unscored — Both systems have positive evidence beyond mere doctrine: measured U.S. uptake and experimentally persuasive PRC content support real but offsetting strengths.
Rival explanations applied
II-A08 Strong: platform and media-system structure, not national capability, determines production and distribution tempo.
II-A09 Strong: message content and policy appeal explain uptake more than agility.
II-A10 Strong: detected covert campaigns are selected failures and may understate more competent undisclosed operations.
II-A11 Strong: independent U.S. cultural and media production is not government-taskable even when it benefits national influence.
Sensitivity and revision conditions
to prc centered edge
Requires matched event evidence of faster accurate production, better localization, and equal or greater priority-audience uptake after bot and paid-placement controls.
to us partner edge
Requires post-2025 matched evidence of superior correction, local fit, and uptake within the relevant decision window.
to insufficient comparative basis
Revert to unscored if measured U.S. audience and experimental PRC persuasion are judged too far downstream and mismatched to inform agility components.
does not move with
- More posts
- More AI-generated items
- More claimed followers
- Faster content without quality or uptake
Fact confidence: Moderate · Comparative confidence: Low · Observability: limited
Return to the ruling →04 · Distribution ReachU.S. / partner edge
RulingThe U.S.-anchored system retains a low-confidence edge. USAGM's FY2024 report provides the only admitted large, method-documented, unduplicated global consumption estimate in this comparison—427 million adults weekly outside the United States, explicitly excluding China and disclosing rolling-survey limits. This official reach sits alongside a vast, independently operated U.S. media, platform, education, and cultural ecosystem and partner channels that are available in specific cases. PRC state media, bureaus, and content-sharing networks are global and can insert material into local ecosystems, but their public totals overwhelmingly describe availability, followers, or activity rather than comparable consumption. The call is narrow because U.S. official media experienced severe 2025-2026 disruption and no current post-disruption matched consumption audit exists.
Strongest counterevidence
- CGTN claims availability in more than 160 countries and more than 600 million followers.
- Xinhua claims 184 overseas bureaus and extensive content supply.
- AidData finds PRC content-sharing partnerships associated with changes in leadership approval in East Asia and Pacific.
- USAGM cannot produce current representative audience data in China and several denied markets.
- USAGM's global total can include country estimates measured up to five years earlier.
- Current U.S. media and program disruption may reduce the continuing value of FY2024 reach.
Why the ruling stops here
- Why not us partner lead — The U.S. total uses rolling surveys, excludes important denied markets, is actor-reported, predates major disruption, and cannot be directly compared with current PRC consumption.
- Why not contested — The U.S. side has positive consumption evidence at the decisive stage, while the PRC side's public global totals remain primarily potential availability and activity rather than a missing-data inference alone.
- Why not prc centered edge or lead — Bureaus, countries, followers, and content-sharing agreements do not establish unduplicated sustained human consumption; detected high-volume campaigns often had negligible engagement.
Rival explanations applied
II-A12 Decisive for confidence: unmatched measurement methods can create an artificial U.S. edge because U.S. institutions publish audience surveys and PRC institutions publish supply counts.
II-A13 Strong: English-language utility and independent commercial culture generate reach that is not government-created or taskable.
II-A14 Strong: censorship and survey restrictions undercount real U.S. reach in China while simultaneously constraining it.
II-A15 Strong: 2025-2026 U.S. institutional disruption may have invalidated the latest measured baseline.
Sensitivity and revision conditions
down to contested
Move to Contested if independent post-2025 U.S. consumption falls materially or if matched PRC state-media and partner consumption approaches it in priority markets.
up to us partner lead
Requires current, independently audited, repeated consumption advantages across consequential regions plus demonstrated resilience after disruption.
to prc centered edge
Requires independently measured, unduplicated PRC consumption above the U.S.-anchored system in a broad matched priority-audience portfolio.
current call not robust to
- A large post-disruption U.S. audience decline
- A credible global PRC audience audit
- Evidence that current U.S. totals substantially double-count or overage consumption
Fact confidence: High · Comparative confidence: Low · Observability: limited
Return to the ruling →05 · Credibility & TrustContested
RulingCredibility and trust are Contested. The most current broad surveys show a major reversal: China now receives higher favorability or leadership approval than the United States in many countries, driven by both improving PRC views and a sharp current U.S. decline. Yet these measures are highly policy- and leader-sensitive and do not measure trust in CGTN, Xinhua, U.S. diplomats, USAGM networks, universities, independent media, or local messengers on matched terms. The U.S. system retains structural credibility assets in independent media, open contestation, correction, cultural institutions, and partners, but current press-freedom and reliability concerns weaken any automatic advantage. PRC central message consistency and improving reputation are positive, while censorship, opaque correction, and very poor information-freedom conditions impose major integrity costs. The evidence supports offsetting positive cases, not an edge.
Strongest counterevidence
- Pew found the U.S. more favorably viewed in six surveyed countries and much more reliable in the Philippines.
- Pew still found broader international belief that the U.S. respects personal freedoms, although the gap narrowed.
- Brand Finance ranked the U.S. first overall, in familiarity, and in influence despite China's reputation lead.
- PRC messages persuaded in one multicountry experiment, but negative wolf-warrior messages backfired in India.
- U.S. press-freedom ranking and domestic news trust have deteriorated, weakening a simple open-system advantage.
- PRC internet and press freedom remain among the world's worst, creating serious integrity costs.
Why the ruling stops here
- Why not prc centered edge or lead — Current country image is not source-specific trust; PRC censorship, opaque correction, and weak press freedom are major counterevidence, and important regional publics still trust the U.S. more.
- Why not us partner edge or lead — Independent media and open institutions cannot erase convergent current survey evidence of declining U.S. reliability, reputation, and leadership approval.
- Why contested not unscored — Matched global opinion surveys and experimental message effects provide positive comparative evidence for both systems even though the source-specific trust component remains unscored.
Rival explanations applied
II-A16 Decisive: foreign policy and leader performance, not information capability, drive much current reputation.
II-A17 Strong: U.S. media pluralism creates visible contradiction and criticism that can reduce simple trust measures while increasing information reliability.
II-A18 Strong: PRC censorship and survey sensitivity can bias measured domestic credibility and conceal correction failures.
II-A19 Strong: trust is audience-, issue-, and messenger-specific; global averages conceal opposite regional results.
Sensitivity and revision conditions
to prc centered edge
Requires current source-specific trust and crisis-reliance measures—not only national or leader favorability—to converge across multiple regions.
to us partner edge
Requires stabilization of current policy effects and matched evidence that independent U.S./partner messengers are relied on more under uncertainty.
to insufficient comparative basis
Use unscored if broad reputation measures are excluded entirely and the remaining source-specific evidence is judged too sparse.
current call robust to
- Treating nation-brand indices as context rather than outcomes
- Separating domestic information freedom from foreign trust
- Regional exceptions in both directions
Fact confidence: High · Comparative confidence: Moderate · Observability: partial
Return to the ruling →06 · Cultural InfluenceU.S. / partner edge
RulingThe U.S.-anchored system has a moderate-confidence cultural-influence edge. Convergent independent evidence shows broader U.S. attraction across arts and entertainment, international education, language use, cultural projection, and several revealed destination preferences. These advantages arise heavily from independent and commercial ecosystems and the global utility of English, so they are national assets but not fully government-taskable. The PRC is narrowing parts of the gap through games, audiovisual products, travel, language learning, and state-supported cultural exports; however, most current PRC evidence stops at supply, revenue, enrollment, or exposure rather than durable affinity or choice. The edge is not a lead because China is growing quickly, current U.S. reputation has deteriorated, and many cultural-choice measures have powerful noncultural causes.
Strongest counterevidence
- China was the largest annual cultural-influence gainer in Lowy's 2025 Asia Power Index.
- Chinese was Duolingo's fastest-growing language in twelve countries and second-fastest in the United States in 2025.
- PRC games, audiovisual content, films, travel, and cultural exports show rapid international growth.
- Pew and Gallup current reputation measures often favor China over the United States.
- ISEAS respondents narrowly preferred the United States over China for holiday and more clearly for living or working, but Japan and regional destinations led both categories.
- New international student enrollment in the United States declined even as the total remained at a record level.
Why the ruling stops here
- Why not us partner lead — PRC cultural supply and language interest are growing quickly; U.S. attraction is partly English utility, market size, and independent production rather than controllable statecraft; current U.S. reputation has weakened.
- Why not contested — Multiple independent measures at or beyond exposure—arts and entertainment reputation, international study, language behavior, and destination preference—converge on a continuing U.S. advantage rather than mere PRC data absence.
- Why not prc centered edge or lead — PRC cultural-export, game, film, travel, and language growth is substantial, but admitted evidence rarely follows exposure into sustained affinity, preference, or behavior across regions.
Rival explanations applied
II-A20 Decisive for attribution: independent U.S. culture, universities, entertainment firms, and English are not state-directed instruments.
II-A21 Strong: study, work, travel, and language choices reflect price, visas, safety, labor markets, capacity, and utility as well as affinity.
II-A22 Strong: global and regional indices use proprietary or expert weights and mix cultural inputs with outcomes.
II-A23 Strong: PRC growth may be undercounted because current independent downstream surveys lag fast-moving cultural products.
Sensitivity and revision conditions
down to contested
Move to Contested if recent PRC cultural growth produces sustained affinity and revealed preference across multiple regions or if U.S. study, language, and entertainment attraction materially declines.
up to us partner lead
Requires causal, longitudinal evidence that cultural contact produces durable affinity and consequential choices across regions after cost and access controls.
to prc centered edge
Requires broad, independently measured reversal in contact-to-affinity and revealed preference, not export volume alone.
current call robust to
- Removing composite soft-power scores
- Treating English utility as a confounder
- Attributing independent culture separately from government capacity
Fact confidence: High · Comparative confidence: Moderate · Observability: partial
Return to the ruling →07 · Influence ConversionInsufficient comparative basis
RulingThe category is explicitly unscored. Rare high-quality studies show that both systems can change short-run attitudes: representative CGTN messages shifted political-economic model preferences in a 19-country experiment; positive PRC public-diplomacy messages produced modest gains in India while combative messages failed; and U.S. embassy democracy messages could improve attitudes inside China under specified conditions. AidData finds associations between PRC content-sharing and leadership approval but no systematic relationship with UN General Assembly voting. These studies establish conditional attitude effects and a serious causal research base, not a representative comparison of consequential decisions, behavior, and persistence. Neither system has a public matched portfolio that clears the full conversion contract.
Strongest counterevidence
- CGTN treatments outperformed sampled U.S. government messages in a preregistered 19-country experiment.
- Positive PRC messaging produced modest attitude gains in India while negative messaging backfired.
- U.S. embassy messaging produced favorable attitude effects inside China under specified conditions.
- PRC content-sharing partnerships were associated with approval changes in East Asia and Pacific.
- The content-sharing study found no systematic UN General Assembly voting alignment.
- No admitted study tracks natural exposure through consequential behavior and persistence on matched terms.
Why the ruling stops here
- Why not contested — The decisive category is conversion into decisions, behavior, and persistence, not attitude treatment effects alone. Positive evidence for both systems does not reach the required stage.
- Why not prc centered edge or lead — PRC experiments and approval associations do not establish natural-exposure behavioral conversion; the admitted UN-vote test found no systematic relationship.
- Why not us partner edge or lead — A U.S. messaging experiment and broad cultural/reputation assets do not establish a matched behavioral conversion advantage.
- Publication boundary — Publish only as an explicit unscored ruling and display the attitude-versus-behavior boundary prominently.
- What would permit a position — A preregistered matched portfolio with natural exposure, consequential decisions or behavior, rival-cause controls, and persistence.
Rival explanations applied
II-A24 Decisive: material incentives, policy performance, coercion, identity, social networks, and preexisting preferences may cause the observed choice rather than information exposure.
II-A25 Strong: forced survey exposure overstates natural reach and attention.
II-A26 Strong: selected messages and countries may not represent either system's campaign portfolio.
II-A27 Strong: publication favors measurable positive effects while failed, stopped, and classified campaigns remain missing.
Sensitivity and revision conditions
to prc centered edge
Requires repeated matched natural-exposure evidence linking PRC information to consequential decisions or behavior with persistent effects and stronger results than U.S. comparators.
to us partner edge
Requires the same evidence for U.S./partner efforts, with independent ecosystem and policy effects separately attributed.
to contested
Requires positive matched decision or behavioral evidence for both systems even if effects differ by issue or region.
does not move with
- Additional followers or impressions
- Country favorability alone
- More survey-treatment attitude effects without behavior
- Policy outcomes without exposure attribution
Fact confidence: High · Comparative confidence: Not assessed · Observability: not assessed
Return to the ruling →08 · Counter-InfluenceContested
RulingThe category is Contested. The U.S.-anchored commercial, research, media, civil-society, and government ecosystem repeatedly attributes and disrupts state-linked operations with public evidence. Google, Meta, Microsoft, and OpenAI reporting provides strong positive detection evidence, and many disclosed PRC-linked campaigns achieved little organic attention. But platform visibility is proprietary and selected, and low attacker engagement cannot be assigned to the defender without a counterfactual. PRC access control can deny foreign messaging at scale, yet severe censorship, press-freedom, false-positive, feedback, and legitimacy costs prevent treating blocking as information integrity. No common attack set gives both systems equal ground truth, exposure, misses, false positives, rights costs, and recovery, so neither edge is defensible.
Strongest counterevidence
- Google reported more than 175,000 lifetime DRAGONBRIDGE channel disruptions and extremely low organic engagement.
- Meta, Microsoft, and OpenAI independently report recurring detection and attribution of state-linked activity.
- Detected operations are selected and platform-specific; undetected campaigns and offline channels are missing.
- PRC blocking limits access to foreign state media and other external content at national scale.
- Freedom House and RSF document severe PRC information-access and press-freedom costs.
- U.S. press-freedom decline, polarization, and platform opacity are serious counterevidence to an uncomplicated open-system lead.
Why the ruling stops here
- Why not us partner edge or lead — Public attributions and takedowns do not measure misses, causal exposure reduction, belief correction, platform incentives, or rights costs; some campaigns were already ineffective.
- Why not prc centered edge or lead — National blocking and censorship can suppress truthful information, impede learning, and impose massive integrity costs; detection accuracy and recovery are opaque.
- Why contested not unscored — Both systems demonstrate positive, consequential capacity in distinct components—transparent distributed detection versus centralized access denial—even though the full outcome contract remains unmatched.
Rival explanations applied
II-A28 Decisive: disclosed campaigns may fail because attackers are weak, not because defenders are effective.
II-A29 Strong: U.S.-based platforms report what they can see and may share institutional incentives, so multiple reports are not fully independent.
II-A30 Strong: PRC censorship can produce a low observed penetration rate while simultaneously degrading truth, trust, and policy feedback.
II-A31 Strong: different legal and normative objectives make takedown, blocking, and integrity measures non-equivalent.
Sensitivity and revision conditions
to us partner edge
Requires a common attack set showing faster accurate detection, greater causal exposure reduction and recovery, fewer false positives, and lower rights costs.
to prc centered edge
Requires the same common-denominator evidence and proof that access control improves belief accuracy and resilience rather than merely suppressing content.
to insufficient comparative basis
Use unscored if transparent detection and centralized denial are judged too different to constitute positive comparative evidence within one capability.
does not move with
- More account removals
- More blocked domains
- Adversary content that already had no audience
- Unverified government attribution
Fact confidence: High · Comparative confidence: Low · Observability: limited
Return to the ruling →09 · Learning & AdaptationInsufficient comparative basis
RulingThe category is explicitly unscored. The U.S. system discloses substantially more evaluation machinery, performance reporting, external program evaluations, oversight findings, and failures. PRC doctrine repeatedly calls for audience tailoring, feedback, localization, and improved communication effectiveness, but public negative findings, stop decisions, and repeated-cycle effect estimates are sparse. The transparency difference is itself important to institutional integrity, yet it cannot establish a comparative learning advantage: public U.S. records rarely complete the chain from finding to changed design to better comparable downstream result, while PRC opacity cannot be coded as nonlearning. No matched repeated-cycle portfolio exists.
Strongest counterevidence
- State and USAGM guidance contains explicit monitoring, evaluation, impact, agility, and data-validation frameworks.
- ECA publishes learning agendas and externally conducted program evaluations.
- USAGM reports multi-year performance trends and measurement limitations.
- GAO and OIG identify recurring coordination, measurement, control, and data-age weaknesses.
- PRC doctrine repeatedly demands audience-specific and effectiveness-oriented international communication.
- Public PRC sources rarely disclose failed campaigns, negative evaluations, or stop decisions.
Why the ruling stops here
- Why not contested — The decisive stages—documented adjustment and better repeated result—lack positive comparable evidence for both systems.
- Why not us partner edge or lead — Greater transparency and more evaluation do not show that findings are used or that comparable outcomes improve; recurring GAO and OIG weaknesses are material.
- Why not prc centered edge or lead — Central direction, localization, and rapid output may reflect adaptation, but doctrine and selected successes do not reveal negative findings, design changes, or repeated gains.
- Publication boundary — Publish only as an explicit unscored ruling; describe transparency separately and never treat opacity as failure or evaluation volume as learning.
- What would permit a position — A matched repeated-cycle portfolio with documented findings, adjustments, failures, and effect-size change.
Rival explanations applied
II-A32 Decisive: openness produces an observability penalty for the U.S. and a success-selection advantage for the PRC.
II-A33 Strong: operational learning may be classified, contractor-held, proprietary, or politically sensitive in both systems.
II-A34 Strong: apparent improvement may result from an easier audience, more resources, policy change, or platform growth rather than learning.
II-A35 Strong: organizational disruption and leadership turnover can destroy institutional memory independently of analytical quality.
Sensitivity and revision conditions
to us partner edge
Requires a representative portfolio tracing findings to design changes and improved repeated results, with failures and nulls included, plus a matched or bounded PRC comparator.
to prc centered edge
Requires comparable, independently reviewable PRC evidence of negative findings, adjustment, repeated gains, and failure retention.
to contested
Requires positive matched repeated-cycle evidence for both systems.
does not move with
- More evaluations
- More dashboards
- An evaluation office
- AI adoption
- Claimed flexibility without repeated results
Fact confidence: High · Comparative confidence: Not assessed · Observability: not assessed
Return to the ruling →Falsifiable hypotheses
The synthesis is conditional.
Twelve strong claims were tested against alternatives, disconfirming evidence, and the capability-level record.
II-H01The U.S. system understands foreign audiences better because it publishes more research, polling, evaluations, and failures.
Strongest alternative
The public difference measures transparency rather than insight accuracy; PRC internal research may be extensive but undisclosed.
What would disconfirm it
A matched portfolio of preregistered forecasts, later validation, documented planning changes, and downstream gains for both systems.
Current adjudication
II-D01 remains Insufficient comparative basis. U.S. transparency is credited as an integrity and observability feature, not as proof of superior audience understanding.
II-H02PRC centralization makes its narrative policy categorically superior.
Strongest alternative
Centralization produces visible consistency by suppressing contradiction and dissent, while policy appeal and censorship—not narrative skill—explain observed effects.
What would disconfirm it
Matched crisis coding showing higher PRC claim-conduct contradiction, concealment, backfire, or failed repair than the U.S. system.
Current adjudication
II-D02 is a low-confidence PRC-centered edge, not a lead. Positive alignment and persuasion evidence survives, but wolf-warrior backfire, integrity costs, and missing contradiction data sharply bound it.
II-H03The system that produces the most multilingual content fastest has superior content agility.
Strongest alternative
High output can be inaccurate, poorly localized, automated, duplicated, or unseen; useful agility requires quality, correction, and priority-audience uptake within the decision window.
What would disconfirm it
Matched event-level latency, localization validation, error correction, and qualified uptake favoring one system.
Current adjudication
II-D03 is Contested. DRAGONBRIDGE is a critical negative case against volume-based judgment.
II-H04CGTN countries, Xinhua bureaus, followers, and PRC content-sharing agreements erase the U.S. reach advantage.
Strongest alternative
The PRC figures measure potential supply and activity, while the U.S. total measures unduplicated weekly human consumption under a documented survey method.
What would disconfirm it
Independent, current, unduplicated PRC consumption in matched priority audiences at or above the U.S.-anchored system.
Current adjudication
II-D04 remains a low-confidence U.S./partner edge. It can fall to Contested quickly if post-2025 U.S. reach declines or a comparable PRC audience series appears.
II-H05China's current global reputation advantage proves superior credibility and trust.
Strongest alternative
Pew and Gallup measure country and leadership evaluation driven heavily by current policy and leaders, not trust in named information sources or messengers.
What would disconfirm it
Matched source-specific trust and crisis-reliance measures that converge with the reputation result across regions.
Current adjudication
II-D05 is Contested. Current reputation is a real PRC strength and operating condition but does not overcome U.S./partner integrity and messenger counterevidence or regional exceptions.
II-H06U.S. cultural popularity proves a durable statecraft lead over China.
Strongest alternative
English utility, commercial scale, independent production, visa and education capacity, cost, and labor-market returns explain much observed attraction; these assets are not fully government-taskable.
What would disconfirm it
Contact-to-affinity and revealed-choice models that remove noncultural drivers and show no U.S. advantage or a PRC advantage across regions.
Current adjudication
II-D06 is a U.S./partner edge, not a lead. Multiple independent and behavioral proxies converge, but attribution and rapid PRC growth prevent a stronger call.
II-H07Experimental attitude shifts demonstrate national-system influence conversion.
Strongest alternative
Forced exposure and selected messages produce short-run survey responses that may not survive natural attention, cost, countermessage, or translate into decisions and behavior.
What would disconfirm it
Matched natural-exposure studies with preregistered objectives, consequential behavior, rival-cause controls, and persistence.
Current adjudication
II-D07 remains Insufficient comparative basis. The experiments are admitted as real attitude-conversion evidence and explicitly prevented from carrying the behavioral category.
II-H08Large platform takedown totals prove a U.S. counter-influence advantage.
Strongest alternative
Platforms may be observing large numbers of already ineffective accounts; totals omit missed operations, false positives, counterfactual exposure, belief correction, and rights costs.
What would disconfirm it
A common attack-set evaluation showing defender-caused exposure reduction and recovery with low false-positive and rights costs.
Current adjudication
II-D08 is Contested. U.S. transparent detection is a component edge, not a full-category edge.
II-H09PRC censorship and blocking prove superior counter-influence because adversary messages cannot penetrate.
Strongest alternative
Suppression can reduce availability while increasing false positives, blocking truthful information, degrading feedback and trust, and imposing legitimacy and rights costs.
What would disconfirm it
A common threat set showing better belief accuracy, lower harmful exposure, recovery, and bounded protected-expression costs under PRC controls.
Current adjudication
Access denial is a PRC component strength but is not coded as information integrity or a category edge.
II-H10More public evaluation and performance reporting proves superior U.S. learning and adaptation.
Strongest alternative
The documents show transparency and measurement capacity, not that findings changed designs or produced better repeated outcomes; PRC internal learning may be undisclosed.
What would disconfirm it
Matched repeated campaign cycles linking valid findings, design changes, failure retention, and improved downstream effect.
Current adjudication
II-D09 remains Insufficient comparative basis. U.S. transparency is separately noted and PRC opacity is not treated as zero learning.
II-H11The four DIME instruments are complete once nine information capabilities receive axis positions.
Strongest alternative
A complete domain can contain explicit unknowns; forcing every capability onto an axis would manufacture precision and corrupt later cross-domain synthesis.
What would disconfirm it
Comparable decisive-stage evidence for Information Understanding, Influence Conversion, and Learning & Adaptation.
Current adjudication
The domain is analytically complete with three directional positions, three Contested positions, and three unscored positions. Completion describes the audit, not universal certainty.
II-H12An overall information-power winner can be inferred by counting directional positions.
Strongest alternative
Capabilities sit at different causal stages, use different units and confidence, and do not substitute for each other. An upstream edge can fail to convert downstream.
What would disconfirm it
A validated causal model showing commensurable weights, substitution, interaction, conversion, and outcome relevance across the nine capabilities.
Current adjudication
No domain composite or winner is calculated. The asymmetric profile is the substantive finding.
20 tests applied before publication.
- Information remains the instrument and influence remains an observed effect; the two terms are never collapsed.
- Every observation is assigned to one causal stage, and downstream stages are never inferred from upstream availability or attention.
- Potential availability, output, languages, posts, bureaus, followers, impressions, exports, and program enrollment remain inputs unless verified humans and effects are observed.
- State-controlled, state-funded independent, partner, commercial, civil-society, cultural, educational, diaspora, and covert contributions retain separate attribution.
- Independent U.S. ecosystem assets count as national informational power only with an explicit non-taskability caveat.
- Partner contributions count only when observed, available, relevant, and timely in the case; alignment labels never pool them automatically.
- Overt persuasion and covert or deceptive activity remain separate, with exposure, effect, attribution risk, legitimacy cost, and persistence recorded.
- Cultural production, export revenue, popularity, awards, enrollment, or audience size establishes exposure, not cultural influence.
- Country favorability, leadership approval, and nation-brand reputation remain contextual and never substitute automatically for source-specific credibility.
- Censorship, blocking, or account removals do not establish counter-influence without detection quality, exposure reduction, false positives, rights cost, and recovery.
- Adversary failure is not credited to a defender without evidence of defender contribution.
- Evaluation volume and transparency do not establish learning without documented adjustment and improved repeated outcomes.
- A weakness, disruption, or missing observation on one side never establishes an advantage for the other.
- Insufficient comparative basis is used instead of Contested when positive evidence for both systems is absent at the decisive stage.
- First-party sources establish doctrine, organization, activity, reported output, or actor claims; they do not independently establish trust, influence, or causal conversion.
- Institutional-family correlation is controlled; citation counts are never treated as independent corroboration.
- Matched claims use the same audience, region, issue, time, denominator, source type, causal stage, and method or disclose each mismatch.
- Global findings survive Indo-Pacific, Europe, Africa, Middle East and North Africa, and Latin America and Caribbean sensitivity checks or are narrowed.
- Current leader and policy effects are separated from durable system capability wherever the data permit.
- No weighted score, position count, or implicit ladder produces an overall information-power winner.