In its end-of-year threat report, Meta says it took down several global disinformation networks, including anti-vaccine groups and groups linked to state actors
The takedowns described in the company's latest threat report demonstrate how the cat-and-mouse game between Facebook and bad actors is escalating
Washington PostElizabeth Dwoskin
Context & Ripple Effects
Meta's report follows its removal of a Nicaraguan government-linked troll farm, showing that its enforcement was aimed at coordinated networks as well as individual posts. The inclusion of anti-vaccine and state-linked operations puts public-health misinformation and geopolitical influence within the same trust-and-safety workflow.
The removed anti-vaccine and state-linked networks lose their Facebook distribution, accounts, and groups, forcing their operators to rebuild audiences or shift activity elsewhere.
Meta must continue identifying coordinated behavior across regions and issue public threat reporting that makes its enforcement posture visible to users and adversaries.
Second-order effects
State-linked influence operators and anti-vaccine organizers face pressure to fragment operations and vary their tactics as Meta's takedowns target networks rather than isolated content.
Facebook's integrity teams become more central to how the platform manages both political influence and health misinformation, broadening the operational burden of enforcement.
Third-order effects
Repeated disruptions point to content moderation becoming an enduring adversarial-security function, with platforms judged on their ability to detect coordinated networks before they scale.
The model is vulnerable to policy reversals: later reporting that Meta dismantled systems for limiting viral hoaxes suggests that enforcement capacity can change even while coordinated influence activity persists.
The trend: Major social platforms are treating coordinated misinformation as a network-security problem, though the durability of that approach depends on sustained investment in integrity systems.
1/ Today we shared our first adversarial threat report, which includes 4 CIB networks, one brigading network, and a mass reporting network that we have removed and are sharing information on. https://about.fb.com/...
Today we're sharing information about attempts to manipulate Twitter - carried out by state-linked actors in six countries - and adding these datasets to our public archive. All associated accounts & content have been removed. https://blog.twitter.com/...
Wilson Edwards, the ‘Swiss biologist’ who did not exist yet wrote a blog undermining efforts to investigate the pandemic origins that was rapidly promoted by Chinese state media. Good to see Beijing's disinformation campaign exposed https://www.vice.com/...
Paging Swiss biologist Wilson Edwards: You've been canceled by Big Tech, dismissed as nothing more than “a hall of mirrors, endlessly reflecting a single fake persona.” https://www.npr.org/...
Meta's new threat report reveals who were behind the initial spreading of the fake Swiss scientist “Wilson Edwards” story - employees of Sichuan Silence Information Technology Co, Ltd 无声信息. https://about.fb.com/... https://twitter.com/...
Remember Wilson Edwards, the alleged Swiss scientist spreading disinformation about Covid's origins? A new Facebook report has confirmed his account was fake (shock) and at the centre of an operation that included a coordinated cluster of Chinese state employees https://twitter.c…
The cat-and-mouse game described in the Facebook's monthly threat report demonstrates how social media is an active battlefield where governments and motivated parties continue to attempt to manipulate public opinion. https://www.seattletimes.com/ ...
Global threats on #disinformation evolve constantly. We strengthen our defenses and add new layers of protection, since threat actors deploy new tactics. Our goal is to make these harmful acts more costly, more difficult to hide and less effective. #meta https://www.washingtonpos…
Hey, Facebook did a good thing! We'll see if they keep doing the good thing once it produces bad press for them, and I do note that it's only *a* good thing and not *all the good things* But still: a good thing! https://twitter.com/...
UPDATE: Twitter just released Information Operation datasets on 8 disclosures including IO actors from — Russia, China, Tanzania, Uganda, Venezuela and Mexico. https://blog.twitter.com/...
It's good to see this transparency initiative from FB. It will help the research community to better understand how influence operations take shape on Facebook. https://twitter.com/...
it's interesting that when Facebook discloses these platform-wide updates on this stuff, what's missing is any indication as to whether any of it ever really mattered https://www.washingtonpost.com/ ... https://twitter.com/...
JUST OUT: Adversarial threat report on brigading, mass reporting and coordinated inauthentic behaviour. With a deep dive into the Chinese operation that created a fake “Swiss biologist” back in July. I think of that one as Operation Swiss Rôle. https://about.fb.com/...
“In Italy and France, an anti-vaccination group known as V_V developed coordinated harassment campaigns against doctors and journalists on Facebook. In some examples, they would call doctors and journalists Nazis for promoting coronavirus vaccines.” https://www.washingtonpost.com…
Weird. I just heard a witness testify to Congress that Facebook had no incentive to keep its site safe. So, I'm going to assume that all of this below is impossible and that Nathaniel is a figment of my imagination. https://twitter.com/...
These FB takedowns are impressive in their own right but this initiative to facilitate research on taken down material is absolutely critical — provides a glimpse into what the backend of a broader research platform (which federal law should compel and broaden) might look like. h…
This is not new for platforms, as Meta also shared today. Most of the activities from countries where threat actors were based seems persistent to disclosures done previously. https://twitter.com/...
This time for China the disclosures include operations that targeted discourse around Xinjiang. Previous set included “of 936 accounts originating from within the People's Republic of China (PRC)” For more information, please refer to this thread: https://blog.twitter.com/...
NEW: Facebook took down a disinformation campaign involving a fake Swiss biologist, which was spread by more than 500 fake accounts. Chinese government officials engaged with the fake post even before the fake accounts did, according to Facebook. https://www.vice.com/...
It is hard to explain and underestimate the importance of this. In every single digital policy class, with every researcher, namely of/on mis- and disinfo, the lack of data was the biggest problem. This is an important step forward. https://twitter.com/...
There's a lot here: * Expanding Crowdtangle IO archive to more researchers * First public takedowns of brigading & mass reporting networks * CIB takedown from Palestine (Hamas) * Two CIB ops focused on Poland / Belarus migrant crisis (one from Belarus KGB) * Op Swiss Rôle
Much to unpack in Facebook's latest report on influence operations, but I especially appreciate the clarity of language. The shift to “Adversarial Threat,” subdivided by methods, is a better way to present a nebulous and difficult phenomenon. 1/ https://about.fb.com/...