The most influential media network operating inside a dozen allied countries today has no masthead, no headquarters and no country of incorporation. It has roughly 50 million members across more than 150 countries. It is the Chinese diaspora — and it is only one of several.
Since the Montevideo Convention of 1933, we have defined a state by four things: a permanent population, a defined territory, a government, and the capacity to conduct relations with other states. Diasporas have the first and, increasingly, the fourth. They skipped the middle two. Territory and government were the parts that made power addressable — the parts you can sanction, summon, deter or simply find on a map.
Governments have borders and rules. Diasporas don’t.
For most of the modern era this was an academic distinction because distance did the work of containment. A community abroad drifted. The second generation lost the language. The third lost the news. Diaspora influence decayed on a predictable curve.
AI switched the curve off. Translation is now instant, free and good enough, which means homeland media reaches anyone of homeland descent regardless of what language they actually speak. Recommendation systems hand every member of a diaspora a personal news service tuned to their hometown, their history, their grievance or their passion. Generative tools manufacture culturally specific content at scale — messages, video, historical narrative, political framing.
Before long, a campaign will not send one message to 500,000 people. It will send 500,000 versions of one message, each optimized for a single person. The third generation will no longer lose the news. The news will find them in a language they never had to learn and in a context they appreciate.
TRUST IS THE ASSET, NOT REACH
What makes this strategically serious is not audience size. Meta and Alphabet have audience size. It is trust — and diaspora trust behaves differently from media trust. Shared identity works as a shortcut: people who are similar to us are presumed credible before they are evaluated. Distance from home raises the value of anyone who still has a connection to it. News about “us” is often processed emotionally, not analytically. And crucially, trust transfers to the sender. We may not extend our belief in a news anchor to the next story on the network, but we do extend our belief in a cousin to whatever the cousin forwards.
That is why interpretation beats reporting. The trusted person who adds forty seconds of commentary to a homeland story carries more weight than the outlet that reported it. It is also why false information moves so efficiently through these networks: forwarding is an act of care. People pass things along to protect each other, and accuracy is rarely the test.
The distribution system is private. Homeland media, host-country media and community media all feed into WhatsApp, WeChat, Telegram, KakaoTalk and Viber — closed groups that are, by design, invisible to nearly every monitoring apparatus we have built.
SIGNAL: READING A DIASPORA’S WIRING
Every diaspora has a signature. Six variables describe it.
S — Self-identity. Who are we, and what creates belonging?
I — Information. How do we understand the world?
G — Group network. How are we connected?
N — Network trust. Whom do we believe?
A — Activation. What moves us to act?
L — Limits and tensions. What divides or constrains us?
Describe those six and you can predict more accurately how a diaspora behaves under pressure. The Ukrainian signature runs on national identity, war-focused information, advocacy networks and trusted veterans and volunteers, activated by threat and solidarity, bounded by war fatigue. The Indian signature runs through professional and alumni networks. The Chinese signature runs through a single application.
That last point deserves more time. WeChat is not a platform the Chinese diaspora uses; it is the infrastructure the diaspora is made of. Payments, news, family, business, community governance — one application, one jurisdiction, one set of rules. When the channel that connects a community is also the channel for reaching it, surveillance, influence and intimidation stop being separate activities.
Russia has demonstrated the adjacent move. Russian-aligned operations have produced fabricated video built to imitate AFP, Deutsche Welle, Euronews and Le Figaro, circulating stories that cast Ukrainian refugees as violent and economically burdensome. Notice the target. The objective was not to persuade the Ukrainian diaspora. It was to shape how host countries perceive it.
And none of it is uniform. Two Russian speakers in the same German city: one consuming state television and Telegram, the other consuming Meduza, Dozhd and YouTube. Same language, same street, two different realities. It is why we must remember that a diaspora is not a monolithic audience. It is contested ground.
WHOSE ACCOUNT IS THIS?
So ask the question that should be uncomfortable: which office in the U.S. government owns this problem?
The most consequential influence terrain of the next decade runs through encrypted private networks in Mandarin, Hindi, Farsi, Russian, Arabic and Punjabi — and institutionally, it is nobody’s account. A narrative can run its full course in those networks and never once surface in English.
A serious response starts with three admissions. First, that diaspora networks are an intelligence and policy subject in their own right, with an owner and a budget line, not a footnote in someone else’s report. Second, that the unit of analysis are the trusted nodes, not the platform — which means people who speak the language and are known in the community. Third, that a diaspora is a constituency whose trust must be earned rather than a target to be managed, because the communities in question can tell the difference immediately and our competitors are already making the offer.
The gap is not only defensive. The United States hosts the largest concentration of diaspora communities on earth: tens of millions of people with language, relationships and standing inside countries where American institutions have almost none. Media flow is two-way — a Ukrainian-American shapes opinion in Ukraine, a Nigerian-American in Lagos. We file that under demographics. Beijing built a permanent apparatus for its version of it.
The leader of tomorrow’s diaspora will not run an organization. They will run a network of trusted nodes. You will not find that person on an org chart, a masthead or a lobbying disclosure.
The map on the wall still shows the world territories. Influence is moving somewhere else. It’s deserving of a new map.
The Cipher Brief is committed to publishing a range of perspectives on national security issues submitted by deeply experienced national security professionals. Opinions expressed are those of the author and do not represent the views or opinions of The Cipher Brief.
Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.
Read more expert-driven national security insights, perspective and analysis in The Cipher Brief
