Evidano is an AI-powered qualitative data analysis platform that accelerates secure transcription, AI-assisted coding, and reproducible synthesis for sensitive research. The June 25, 2026 House Select Committee on China hearing documented tactics from United Front access to coercion and incentives at the subnational level. This post shows how researchers and policy teams can convert a hearing transcript into actionable themes, segment comparisons, and stakeholder-ready evidence using a repeatable qualitative workflow and secure tooling.
Key Takeaways
Key takeaways: The June 25, 2026 hearing framed a multi-vector subnational influence campaign, including access, coercion, and incentives, with witnesses citing annual transfers in the hundreds of billions. Qualitative research workflows that combine defensible coding, cross-segment comparison, and PII redaction can convert long, noisy transcripts into briefing-ready evidence within 48 hours.
- The June 25, 2026 hearing described influence levers as Access, Coercion, Incentives, and Prepositioning, and witnesses referenced annual transfers in the hundreds of billions.
- A reproducible qualitative workflow should include a locked codebook (10–12 codes), frequency counts by speaker/segment, co-occurrence networks, and an immutable audit trail for oversight.
- Secure processing steps include encrypted storage, private LLMs with no third-party training, and PII redaction prior to stakeholder reporting.
- A two-day pilot can ingest video/transcript, auto-suggest codes, run cross-segment analysis, and export redacted, clickable quotes for decision-makers.
Fast take + source
Fast take: The June 25, 2026 hearing framed the CCP’s campaign as a multi-vector operation targeting state and local institutions, and witnesses estimated annual wealth and knowledge transfers in the hundreds of billions. Read the hearing summary at Small Wars Journal.
- Why this matters to qualitative researchers: long transcripts, overlapping claims, and political sensitivity demand accurate theme extraction, segment comparison, and defensible reporting.
- Payoff: within a single day you can ingest the hearing video/transcript, extract themes (Access, Coercion, Incentives, Prepositioning), quantify frequency by speaker/section, and deliver redacted quotes for decision-makers.
Findings snapshot
Findings snapshot: The table below summarizes key metrics and sources cited during the hearing and related statements.
Findings snapshot
| Date / Item | Metric / Fact | Value | Source / Note |
|---|---|---|---|
| June 25, 2026 hearing | Event | House Select Committee on China, public testimony | Small Wars Journal (hearing summary) |
| Economic impact | Estimated annual transfer (Made in China 2025 impacts) | $600 billion | Testimony (David Shedd) |
| IP loss estimates | Range cited in hearing | $225–$600 billion annually | Hearing Q&A |
| Research community | Share of top AI researchers of Chinese origin | 38% | Ranking member statement |
| STEM pipeline | Share of STEM grads from outside U.S. | 72% | Ranking member statement |
What happened (plain English)
What happened: The hearing documented how the CCP’s United Front and state-directed programs operate at the subnational level using three levers: access, coercion, and incentives, and witnesses described physical prepositioning and the use of U.S. nonprofits for influence. The hearing described access via trade delegations, sister cities, and university partnerships, coercion via family ties and business dependencies, and incentives via commercial or political benefits.
- Scope: state and local governments, public universities, school boards, local businesses, and diaspora groups.
- Key allegation: scale and sophistication have accelerated over the last decade, with some estimates placing losses in the hundreds of billions annually.
- Policy tension highlighted: strengthen counterintelligence while avoiding ethnicity-based profiling, civil-rights concerns were emphasized.
Implications for researchers and analysts
For qualitative researchers & UX teams
Qualitative researchers and UX teams should expect long, multi-speaker transcripts with overlapping claims and emotionally loaded language, and their priorities are to build a defensible codebook, quantify theme frequency by speaker and segment, and produce redacted, attributable quotes for stakeholders.
Practical risk: mislabeling a quote or conflating a policy critique with ethnic profiling undermines credibility, so researchers should use audit trails and source tagging.
For university admins & research security
University administrators and research security teams should identify research topics and grants that map to targeted sectors such as AI, biotech, and telecom, and they should track collaboration patterns and flag partnerships that lack disclosure or conflict-of-interest documentation.
Operational step: combine qualitative findings, which surface narrative patterns, with administrative data such as grant recipients and foreign funding for cross-validation.
For policy and oversight teams
Policy and oversight teams should use qualitative synthesis to surface recurring influence channels such as tax-exempt organizations, land purchases, and sister-city deals, and prioritize investigations with triangulated evidence.
Remember: craft policies based on conduct and documented evidence, not ancestry; document analysis choices and safeguards to withstand civil-liberties review.
Do more, faster with Evidano (mapped to this hearing)
Problem: Massive hearing transcript → Solution: Fast ingestion & thematic coding
Problem: Massive hearing transcripts are difficult to process rapidly, Solution: Evidano ingests video and transcript files, auto-transcribes with a custom dictionary for names, agencies, and technical terms, timestamps content, and auto-suggests initial codes from the corpus.
Output: code frequency by speaker and a theme timeline that investigators can use the same day.
Problem: Mixed languages & sensitive PII → Solution: Translation + PII redaction
Problem: Mixed languages and sensitive PII complicate sharing, Solution: Evidano’s translation with custom dictionaries preserves technical meaning, and PII redaction masks sensitive identities for shared reports.
Problem: Need defensible comparisons → Solution: Cross-segment analysis & audit trail
Problem: Comparing themes across segments must be defensible, Solution: Evidano compares themes across segments such as witness type or geography and exports an audit trail showing who applied or adjusted codes, useful for reproducible oversight and civil-liberties review.
Problem: Stakeholders want quick evidence → Solution: Clickable quotes, co-occurrence networks & visuals
Problem: Stakeholders want concise evidence, Solution: Evidano generates clickable quotes, co-occurrence networks to map links like Access with universities or land purchases, and hierarchical code→subcode visualizations for briefings.
Contextual link: Evidano.
Security & ethics
Security and ethics: Evidano uses encrypted storage and private LLMs tuned for qualitative research, and data is never used to train third-party models, which is important when handling politically sensitive hearings.
Tip: pair evidence-focused coding with bias checks and document anti-profiling safeguards in your final memo.
Checklist: Reproduce this analysis (2-day pilot)
Checklist: To reproduce this analysis in a two-day pilot, follow the steps below.
- Step 1: Gather artifacts: recording, caption file, witness bios, and related filings.
- Step 2: Ingest into Evidano: upload video/transcript; run auto-transcription with a custom dictionary for names (e.g., Eileen Wang) and acronyms (MSS, CFIUS).
- Step 3: Auto-code pass: let the model suggest themes (Access, Coercion, Incentives, Prepositioning); review and lock a 10–12 code codebook.
- Step 4: Run cross-segment analysis: compare themes by speaker type (witness, member, staff) and by time segments (opening, testimony, Q&A).
- Step 5: Validate: spot-check 10–20 coded excerpts; run a co-occurrence network to surface hidden linkages such as nonprofits and data centers.
- Step 6: Redact & export: apply PII redaction, export an executive deck with clickable source links and visuals.
- Step 7: Archive & audit: store the project with an immutable audit trail for oversight and response to civil-liberties queries.
FAQ: Subnational influence qualitative analysis
What is qualitative analysis of subnational influence and when should I use it?
Answer: Qualitative analysis of subnational influence is the systematic coding and synthesis of spoken and written records to identify recurring influence channels, tactics, and actors. Use it when you need evidence-based narrative linked to source excerpts for policy, compliance, or oversight action.
How can I compare segments reliably, for example state versus federal concerns?
Answer: You can compare segments reliably by defining consistent segment tags such as speaker role, geography, and institution, then running frequency and cross-segment analyses. Evidano automates tagging and quantifies theme prevalence with exportable tables and visuals.
How secure is AI-enabled research with politically sensitive data?
Answer: AI-enabled research can be secure if you use platforms that provide encryption, private LLMs, data non-training guarantees, and PII redaction, and if you document consent and retain audit logs to support ethical and legal review.
Wrapping up & next steps
Wrapping up: The June 25, 2026 hearing is a rich case for applied qualitative research because it contains layered claims, policy trade-offs, and civil-liberties stakes. The right workflow turns that noisy record into quantified themes and defensible evidence for policymakers.
- Your next two moves: (1) Pull the transcript and tag speakers; (2) Run an AI-assisted code pass and produce a short brief with redacted quotes.
- Ready to try it? Start a pilot and Try Evidano for free to import the hearing transcript, run thematic and cross-segment analyses, and export stakeholder-ready reports within 48 hours.