View profile

Why 40,000 fake Poles should worry Facebook

Revue
 
Here's a report from the University of Oxford about what the authors call "computational propaganda":
 
December 6 · Issue #41 · View online
The Interface
Here’s a report from the University of Oxford about what the authors call “computational propaganda”: “automated social media bots, organized networks of fake online identities, and coordinated trolling campaigns that have become increasingly prevalent and are rapidly being established as an important aspect of contemporary digital politics.”
The authors are interested in Poland, which was formerly considered a postwar democratic success story and has lately lurched rightward. This has happened during a time when most of the population has finally gotten online — and, as you might expect, gotten on Facebook. 
This has led to the creation of a new Polish industry devoted to the creation of fake online citizens who can be deployed in the service of misinformation
Forgive the long excerpt, but this is wild.
Over the past ten years, his firm (which we’ll refer to here as “The Firm”) created more than 40 thousand unique identities, each with multiple accounts on various social media platforms and portals, a unique IP address, and even its own personality, forming a universe of several hundred thousand specific fake accounts that have been used in Polish politics and multiple elections (Daedalus, personal correspondence, 14/01/17).
The process begins with a client: a company in the private sector (pharmaceuticals, natural resources), or a political party/campaign. A strategic objective is outlined and a contract that includes “word of mouth” or “guerrilla” marketing services is written up. An employee of The Firm then starts by creating an email address via a large provider (such as Gmail). Using this email and an invented name, they create accounts on multiple platforms and portals. A suitable profile photo is found via an image search and modified in Photoshop so that it will not appear in a Google image search, and the employee begins posting on various platforms and building a comment history. Each employee manages up to 15 identities at a time, with each having a coherent writing style, interests, and personality. They use a modified VPN to spoof IP addresses so that their accounts will have a series of associated addresses, allowing them to post from multiple locations in a predictable way (as would befit a normal user using a mobile phone and travelling around a city, or using their laptop from home/work/elsewhere).
When these accounts are ready to begin posting on comment sections and Facebook groups or pages, the employee uses only unique content (each account never copies or repopulates posts) as to make it unsearchable and difficult to link to other accounts. All steps are taken so that these accounts are very difficult (in the words of the research subject, “completely impossible”) to conclusively identify as fake.
This all provides a level of deniability for the client, who may not even know exactly (and probably does not want to know) what techniques are being used by their marketing consultants. Furthermore, this is a low risk endeavor: while these processes violate the terms of service for platforms, they exist in a legal grey area. If a firm takes the basic precautions described above, it is highly unlikely that this activity will ever be exposed, and if it is, it is not clear how either the firm or their clients would legally be held accountable.
It’s a vexing problem. Inauthentic accounts backed by motivated human beings, often with the support of the state, may be nimble enough to evade detection of platforms like Facebook and Twitter. And while Facebook in particular invests in removing these accounts, the cat-and-mouse is growing ever more sophisticated. 
A question I had reading this study was the extent to which they are already being used in America. It’s certainly something to keep an eye on as we head into 2018.

Democracy
Conservative Weekly Standard to aid in Facebook fact-checks, prompting outcry
Entering the Facebook ecosystem at age 6
Facebook’s Algorithm Hijacked This $8 Billion Company to Sell Cat Blindfolds
Inside Oracle’s cloak-and-dagger political war with Google
Elsewhere
Glassdoor: Facebook is consistently the best tech company to work for in the U.S.
HQ Trivia is raising money in a round that could value its company at $100M
Launches
The forthcoming Vine sequel has a name and a logo: V2
Takes
News Corp CEO: "YouTube Is a Toxic Waste Dump"
And finally ...
I Made My Shed the Top Rated Restaurant On TripAdvisor
Talk to me
Questions? Comments? Fake TripAdvisor reviews? casey@theverge.com
Did you enjoy this issue?
Thumbs up 1ae5a7bdfcd3220e2b376aa0c1607bc5edaba758e5dd83b482d03965219a220b Thumbs down e13779fa29e2935b47488fb8f82977fedcf689a0cc0cc3c19fa3c6bb14d1493b
If you don't want these updates anymore, please unsubscribe here
If you were forwarded this newsletter and you like it, you can subscribe here
Powered by Revue