The Deceptors

“During occasions when new industries and technologies are developed, the physical and physiological details usually taken as given can become a matter of concern with consequent clarification of the assumptions and conceptions we have of what individuals are.” (p.7)

Written in 1969, this quotation from the sociologist Erving Goffman’s book Strategic Interaction retains striking contemporary resonances. First, is the recognition that there is a recurring historical pattern of significant new technologies inducing profound disruptions to the social order. And then, associated with which, are shifts in the construction of what being an individual is understood to mean and involve in such settings. In the information age, shaped by and shaping of its core technologies of the internet and social media, there is rampant public and political consternation about how these technologies, and the behaviours they enable and encourage, are contributing to an increasingly polarised societal order. 

When one is able to obscure who they really are, different forms of behaviour become more viable.

One specific form of digital behaviour worth paying particular attention to is the normalisation of various forms of false persona and fabricated identity. For when one is able to obscure or obfuscate who they ‘really’ are, different forms of behaviour, often of the more problematic and troubling kinds, become more viable and plausible. Anonymity (literally being unknown) or pseudo-anonymity (assuming a false identity) provide the basis for what Thomas DeGloma labels ‘subversive cover identities'.

These are part of the standard ‘toolkit’ for the design and delivery of state-linked information manipulation operations, where operators design and deploy various cover identities to deliver disinforming, distorting, and deceptive messaging. These messages are intended to impact public understanding and political decision-making by their perceived adversaries, and sometimes allies as well. The archetype for this kind of influencing work was the St Petersburg based Internet Research Agency. They originally acquired international public notoriety having been discovered trying to covertly shape the outcomes of the 2016 US Presidential election, using large numbers of fake social media accounts with diverse cover identities to camouflage their actions and motivations. 

Recently however, researchers have been afforded novel and innovative insights into how this kind of malign, remote influencing work is organised and conducted. In Autumn 2024, a leak of over 2,700 internal documents from a Russian organisation called the Social Design Agency (SDA) were made available to investigative journalists and researchers. These provide unprecedented and exquisite details about the range of digital influence engineering behaviours the SDA operators were using. Informed by these materials, below are highlighted two deceptive digital behaviours that deserve more attention.

Comment Manipulation

To date, most studies of state information manipulation efforts have focused upon cataloguing and describing the use of deceptive account cover identities, the dissemination of disinforming and/or distorting narratives and images, the deployment of large numbers of bots to spread their messages, and similar, across mainstream social media platforms. However, the SDA leak dataset reveals that their operators invested significant time, effort and resource in posting fake ‘reader comments’ under news stories aligned with their campaigns and interests. For example, in a period of 5 months in 2022, SDA operatives recorded posting 14,606 comments across 21 UK media outlets. This included national media such as the Daily Mail website, Guardian, and BBC. But also, more intriguingly, a number of more locally oriented publications such as Wales Online, the Manchester Evening News, and Liverpool Echo. 

A key objective for the operatives is to seed conflict, antagonism, and division.

We can infer several motivations for them using this particular tactic. First, it is far less likely to get detected than posting original content on social media platforms, and involves far less effort than fabricating and curating a full cover identity. Second, given that a key objective for the operatives is to seed conflict, antagonism, and division, posting polarising messages in this way is fully aligned with their purposes. Indeed, the data in the leaked files suggest they have sequences of pre-prepared comments that can be posted to confect the impression of online conflict and contestation. 

Fake Polls and Sociological Research

A second more innovative technique detailed within the documents is the use of fake public opinion polls. These were used as a creative way of getting politically contentious and divisive statements in front of people recruited under the guise that they were respondents in a legitimate research effort, on the pretence of asking them to either endorse or refute the claim made. But the real purpose was to have them engage with some highly partisan or distorted statement, that if encountered in some other context (such as on a social media platform) many people might very well ignore.

Again, as with the previous example, the scale of exposures achieved via this route is considerable. Of further note, is that the SDA also conducted focus group research in Kazakhstan, engaging directly with citizens to collect in-depth qualitative accounts on a number of potentially exploitable public concerns.

The purpose of this article has been to start to map out more fully the range of deceptive digital influence engineering behaviours used by state actors, engaged in organising and conducting information influence operations. For whilst there is abundant societal concern about mis- and disinformation, the sustained and systematic nature of the deception that state agencies and their proxies can leverage warrants particular attention.