Concerned you’re being duped by a bot or troll account online? Then you’ll want to keep reading this guide where we detail the top tools and tips to help you spot inauthentic and automated accounts on major social networks.
NordVPN £2.56 + 3 months
NordVPN is now offering up 66% off along with 3 months free using the code TrustedReviews at checkout!
- NordVPN
- CODE: TrustedReviews
- from £2.56/mo
Twitter trolls
Of all social networks, Twitter and its powerful API (application programming interface) provide the easiest access and data needed for robust detection of suspicious accounts, but not all tools are equal on this front.
Many “free” Twitter follower auditing tools are just trying to sell you their full reporting service. They’ll ask you to add their app to your account, but most won’t provide a list of your “fake” followers free of charge, and they often include inactive accounts in the figures they produce – a far cry from fakery.
However, the Indiana University Observatory on Social Media’s Botometer is on the level and clearly distinguishes between different kinds of potentially undesirable account, although it will take a while to work through all your followers if you have a lot of them. Fortunately, you can check specific accounts, too.
Also useful is Bot Sentinel, a community project that stores and publishes results about accounts as people check them, creating a valuable public archive of problematic accounts.
Analysing Reddit and Facebook
Other major social networks are less forthcoming with regards to coughing up data that could be used to analysis the authenticity or otherwise of the people you interact with. However, there’s a significant body of research about troll and spam accounts that give us some general rules of thumb for making judgements.
The TROLLMAGNIFIER project looked at methods of detecting state-sponsored troll accounts on Reddit. No publicly-available software tools have yet come of it, but the research team found that state-sponsored troll accounts in particular tended to loosely coordinate with one another, interacting to push specific narratives and manufacture conflicts – performative augments – between one another about the topics they wished to propagandize.
Facebook publishes regular reports on such “coordinated inauthentic behaviour”, with a focus on not only state-sponsored trolls, but also financial scams and targeted social engineering attacks designed to get prominent influencers and journalists to legitimise the scammers.
Facebook’s 2021 threat report indicates that coordinated trolls increasingly have specific targets in mind for their “influence operations”, defined as “coordinated efforts to manipulate or corrupt public debate for a strategic goal.”
Tiktok and Instagram fakery
Inauthentic accounts are rife on TikTok and Instagram, with a bustling trade in click-farmed likes. A current popular rule of thumb for spotting accounts that buy follower and attention is to quickly view an influencer’s followers to see how many have filled out profiles with avatars, but you’ll generally have to do this by hand. Social marketing firm HypeAuditor provides fake follower audits for TikTok and Instagram, but these provide very few free searches per IP address (use a VPN if you need more) and will only show you analysis of accounts already in their database, which appears to include only the biggest influences.
More usefully, when it comes to spotting propaganda on TikTok, Tracking Exposed has turned its analysis tools to the task of spotting it, publishing a detailed report on Russian military propaganda in April 2022. Tracking Exposed has tools available to help analyse TikTok, YouTube, and Facebook accounts, but, aimed at academics gathering data en masse, the metadata they spit out doesn’t readily lend itself to fast assessment of potential fakery by the layperson.
NordVPN £2.56 + 3 months
NordVPN is now offering up 66% off along with 3 months free using the code TrustedReviews at checkout!
- NordVPN
- CODE: TrustedReviews
- from £2.56/mo
Spotting the scammers
As with much of online safety, such as spotting phishing attempts, identifying inauthentic accounts requires us to keep in mind that not everything is as it seems on the internet.
- Consider who is running the account – on Facebook, pages will have identified managers and groups have admins. Where are they based? A US politics page run by someone from North Macedonia may not be all it seems.
- Has the account always posted this kind of content? YouTube accounts with large followings are bought and sold, and you can sometimes see that the older content was made by entirely different people than those now running the channel.
- Has the account been verified by the social media platform? This only really applies if the account purports to be an official body or influential figure. Watch out for verified accounts being stolen and renamed or used to post out-of-character content, such as when verified Twitter accounts are taken over by cryptocurrency scammers.
You might like…
Editorial independence
Editorial independence means being able to give an unbiased verdict about a product or company, with the avoidance of conflicts of interest. To ensure this is possible, every member of the editorial staff follows a clear code of conduct.
Professional conduct
We also expect our journalists to follow clear ethical standards in their work. Our staff members must strive for honesty and accuracy in everything they do. We follow the IPSO Editors’ code of practice to underpin these standards.
Editorial independence
Editorial independence means being able to give an unbiased verdict about a product or company, with the avoidance of conflicts of interest. To ensure this is possible, every member of the editorial staff follows a clear code of conduct.
Professional conduct
We also expect our journalists to follow clear ethical standards in their work. Our staff members must strive for honesty and accuracy in everything they do. We follow the IPSO Editors’ code of practice to underpin these standards.