Moscow’s Mission: Fast Shares, No Checks, Endless Fakes
How manipulation moves faster than proof
Voice & Vision | Moscow’s Mission, Part 5 | Internet speed, synthetic media, and trust
The playbook didn’t need your vote to change. It only needed your feed to speed up.
False and sensational claims travel farther and faster than true ones on large platforms because novelty and emotion outperform nuance in the click economy. That means an attacker can seed a few sharp stories and let us do the rest of the damage. (Science 2018)
Speed and scale collide with human wiring.
We like information that flatters our identities and confirms what our group already believes. That is why engagement cues, likes, shares, and views, can trick us into trusting what is popular over what is proven. When a claim appears to be everywhere, it starts to feel important. When it feels important, people share it faster. The feed turns attention into evidence, and that is exactly the environment manipulators depend on. (HKS Misinformation Review 2020)
Platforms tried to slow the fire with labels, friction, and takedowns. The results are mixed. Large experiments during the 2020 cycle found that changing what people see in feeds did not flip partisan views in big, immediate ways. Yet the same research showed how resharing pipelines are closely tied to the spread of low-quality claims. Translation: the big levers may not move minds quickly, but the share button can still move rumors fast. (Science 2023 Facebook experiments)
On top of that, the recommender era keeps shifting the ground under our feet. Independent audits and observational studies point to two hard facts for defenders. First, low-credibility content can gain extra lift inside algorithmic systems built to maximize attention. Second, even when a recommender funnels users into denser networks, the echo from familiar voices keeps the appetite for easy claims alive. (arXiv 2023; arXiv 2024)
Now add cheap synthesis.
Fabricated audio, video, and images lower the cost of flooding the zone and raise the burden on every editor, official, and citizen. U.S. agencies warned in 2024 that foreign actors were already using manufactured videos to amplify domestic narratives and cast doubt on election processes. The point is not only to persuade. It is to pollute, so people stop believing anything can be reliably known. (FBI-CISA-ODNI 2024)
The public feels the fatigue. Years of exposure have left Americans saying made-up news is a major national problem, one that erodes trust in institutions and in one another. Concern about AI-generated news and visuals also climbed in 2024 across democracies, which means audiences are bracing for a flood, not a fix. (Pew 2019; Reuters Institute 2024)
This is why attackers hold a structural edge.
They need one hit in a million. Defenders have to be right all the time, across languages, platforms, formats, and time zones. Attribution rarely undoes impact. Even when a platform identifies the source, the screenshot lives on, the narrative migrates to native accounts, and the argument becomes “what everyone is saying.” (Science 2018; Science 2023)
What helps in the real world is not one silver bullet, but layers. Slow the spread of unverified virals. Require provenance for political media buys. Invest in fast forensic teams for newsrooms and states. Teach simple, repeatable pause-and-check habits that average users will actually use. Keep expectations sober, because there is no herd immunity to manipulation. There is only resilience that buys time and preserves trust. (HKS Misinformation Review 2020; Reuters Institute 2024)
That is the permanent problem of manipulation at internet speed. The attacker does not need to win every argument. They only need to make truth slower, trust weaker, and the public more exhausted.
Part 6 turns from the information fight to the civic one. Because the deeper question is not only how we stop the next fake from spreading. It is how communities rebuild shared reality when the incentive structure rewards the opposite.
Sources Vosoughi, Roy, and Aral, “The Spread of True and False News Online,” Science (2018).
Guess, Nyhan, Reifler, et al., 2020 Facebook election experiments, Science (2023).
Roozenbeek et al., “Fake News Game Confers Psychological Resistance,” HKS Misinformation Review (2020).
Corsi, “Evaluating Twitter’s Algorithmic Amplification of Low-Credibility Content,” arXiv (2023).
Duskin et al., “Echo Chambers in the Age of Algorithms,” arXiv (2024).
FBI Press Room, “FBI Warns Public of Artificially Generated Content,” and “FBI, ODNI, and CISA Joint Statement on Election Security” (2024).
Pew Research Center, “Many Americans Say Made-Up News Is a Critical Problem for the Country” (2019).
Reuters Institute for the Study of Journalism, Digital News Report 2024, sections on AI and trust.
