Showing posts with label META. Show all posts

Saturday, 29 March 2025

Techniques for suppressing health experts' social media accounts, part 1 - The Science™ versus key opinion leaders challenging the COVID-19 narrative

Written for researchers and others interested in the many techniques used to suppress COVID-19 dissidents' social media accounts, and digital voices.

There has been extensive censorship of legitimate, expert criticism during the COVID-19 event (Kheriaty, 2022. Shir-Raz et al, 2023. Hughes, 2024). Such scientific suppression makes the narrow frame visible for what the sponsors of global health authoritarianism permit for questioning of The Science™. In contrast to genuine science which innovates through critique, incorporated science does not welcome questioning. Like fascism, corporatist science views critiques of its interventions to be heresy. In the COVID-19 event, key opinion leaders who criticised the lack of scientific rigour behind public health measures (such as genetic vaccine mandates) were treated as heretics by a contemporary version of the Inquisition (Malone et al., 2024). Dissidents were accused of sharing "MDM" (Misinformation, Disinformation and Malinformation) assumed to place the public's lives at risk. Particularly in prestigious medical universities, questioning the dictates of health authorities and their powerful sponsors was viewed as being unacceptable, completely outside an Overton Window that had become far more restrictive due to fear- mongering around a "pandemic" (see Figure 1).

Figure 1. Narrowed Overton Window for COVID-19. Figures copied from (p137-138) in Dr Joseph Fraiman (2023). The dangers of self-censorship during the COVID-19 pandemic. In R. Malone, E. Dowd, & G. Fareed (Eds.), Canary In a Covid World: How Propaganda and Censorship Changed Our (My) World (pp. 132-147). Amazon Digital Services LLC - Kdp.

Higher Education is particularly susceptible to this groupthink, as it lends itself to a purity spiral, which in turn contributes to the growing spiral of silence for "unacceptable views". A purity spiral is a form of groupthink in which it is more beneficial to hold some views than to not hold them. In a process of moral outbidding, individual academics with more extreme views are rewarded. This was evidenced at universities where genetic vaccine proponents loudly supported the mandatory vaccination of students, despite them having minimal, if any, risk. In contrast, scholars expressing moderation, doubt or nuance faced ostracism as "anti-vaxxers". In universities, there are strong social conformity factors within its tight-knit community. Grants, career-support and other forms of institutional support depend on collegiality and alignment with prevailing norms. Being labeled a contrarian for questioning a ‘sacred cow’, such as "safe and effective" genetic vaccines, is likely to jeopardise one’s reputation, and academic future. Academic disciplines coalesce around shared paradigms and axiomatic truths, routinely amplifying groupthink. Challenging reified understandings as shibboleths can lead to exclusion from conferences, journals and cost scholars departmental, faculty, and even university support. Particularly where powerful funders object to such dissent!

Here, administrative orthodoxy can signal an “official” position for the university that chills debate. Dissenters fears of isolation and reprisal (such as poor evaluations and formal complaints for not following the official line) may convince them to self-censor. Particularly where the nonconformist assesses that the strength of opinion against his or her opinion is virulent, alongside high costs to expressing a disagreeable viewpoint- such as negotiating cancelation culture. Individuals who calculate that they have a low chance of success to convince others, and are likely to pay a steep price, self censor and contribute to the growing spiral of silence. The COVID-19 event serves as an excellent example for this growing spiral’s chilling effect versus free speech and independent enquiry.

COVID-19 is highly pertinent for critiquing censorship in the Medical and Health Sciences. Particularly as it featured conflicts of interest that contributed to global health "authorities" policy guidance. Notably, the World Health Organisation promoted poorly substantiated and even unscientific guidelines (Noakes et al., 2021), that merit being considered MDM. In following such dictates from the top policy makers of the Global Public-Private Partnership (GPPP or G3P), most governments' health authorities seemed to ignore key facts. Notably: i. COVID-19 risk was steeply age-stratified (Verity et al, 2019. Ho et al, 2020. Bergman et al, 2021); ii. Prior COVID-19 infection can provide substantial immunity (Nattrass et al., 2021); iii. COVID-19 genetic vaccines did not stop disease transmission (Eyre et al. 2022, Wilder-Smith, 2022); iv. mass-masking was ineffective (Jefferson et al., 2023. Halperin, 2024); v. school closures were unwarranted (Wu et al., 2021); and, vi. there were better alternatives to lengthy, whole-society lockdowns (Coccia, 2021, Gandhi and Venkatesh, 2021. Herby et al., 2024). Both international policy makers' and local health authorities' flawed guidance must be open debate and rigorous critique. If public health interventions had been adapted to such key facts during the COVID-19 event, the resultant revised guidance could well have contributed to better social-, health-, and economic outcomes for billions of people!

This post focuses on six types of suppression techniques that were used against dissenting accounts whose voices are deemed illegitimate "disinformation" spreaders by the Global public-Private Partnerships (G3P)-sponsored industrial censorship complex. This an important concern, since claims that the suppression of free speech's digital reach can "protect public safety" were proved false during COVID-19. A case in point is the censorship of criticism against employee's vaccine mandates. North American employers' mandates are directly linked to excess disabilities and deaths for hundreds and thousands of working-age employees (Dowd, 2024). Deceptive censorship of individuals' reports of vaccine injuries as "malinformation", or automatically-labelling criticism of Operation Warp Speed as "disinformation", would hamper US employee's abilities to make fully-informed decisions on the safety of genetic vaccines. Such deleterious censorship must be critically examined by academics. In contrast, 'Disinformation-for-hire' scholars (Harsin, 2024) will no doubt remain safely ensconced behind their profitable MDM blinkers.

This post is the first in a series that spotlights the myriad of account suppression techniques that exist. For each, examples of censorship against health experts' opinions are provided. Hopefully, readers can then better appreciate the asymmetric struggle that dissidents face when their accounts are targeted by the censorship industrial complex with a myriad of these strategies spanning multiple social media platforms:

Practices for @Account suppression

#1 Deception - users are not alerted to unconstitutional limitations on their free speech

Social media users might assume that their constitutional right to free speech as citizens will be protected within, and across, digital platforms. However, global platforms may not support such rights in practice. No social media company openly discloses the extent to which users' accounts have, and are, being censored for expressing opinions on controversial topics. Nor do these platforms explicitly warn users what they consider to be impermissible opinions. Consequently, their users are not be forewarned regarding what may result in censorship. For example, many COVID19 dissidents were surprised that their legitimate critiques could result in account suspensions and bans (Shir-Raz, 2022). Typically, such censorship was justified by Facebook, Google, LinkedIn, Tik Tok, Twitter and YouTube, due to users' violation of "community rules". In most countries, the freedom of speech is a citizen’s constitutional right that should be illegal to over-ride. It should be deeply concerning that such protections were not supported in the Fourth Estate of the digital public square during the COVID-19 event. Instead, the supra-national interests of health authoritarians came to supersede national laws to prevent (unproven) harms. This pattern of censorship is noticeable in many other scientific issue arenas, ranging from criticism against man-made climate change to skeptics challenging transgender medical ideology.

#2 Cyberstalking - facilitating the virtual and physical targeting of dissidents

An individual who exercises his or her voice against official COVID-19 narratives can expect to receive both legitimate, pro-social and unfair, anti-social criticism. While cyberstalking should be illegal, social media platforms readily facilitate the stalking and cyber-harassment of dissidents. An extreme example of this was Dr Christine Cotton's experiences on LinkedIn. Dr Cotton was an early whistleblower (Jan, 2022) against Pfizer's COVID-19 clinical trial's false claims of 95% efficacy for its treatments.

Her report identified the presence of bias and major deviations from good clinical practice. In press interviews, she reported that the trial did ‘not support validity in terms of efficacy, immunogenicity and tolerance of the results provided in the various Pfizer clinical reports that were examined in the emergency by the various health authorities. Christine shared this report with her professional network on LinkedIn, asking for feedback from former contacts in the pharmaceutical industry. The reception was mostly positive, but it and related posts were subject to a rapid content takedown by LinkedIn, ostensibly for not meeting community standards. At the same time, her profile became hypersurveiled. It attracted unexpected visits from 133 lawyers, the Ministry of Defence, employees of the US Department of State, the World Health Organisation, and others (p142). None of these profile viewers contacted her directly.

#3 Othering - enabling public character assassination via cyber smears

Othering is a process whereby individuals or groups are defined, labeled or targeted as not fitting in within the norms of a social group. This influences how people perceive and treat those who are viewed as being part of the in-group, versus those in an out-group. At a small scale, othering can result in a scholar being ostracised from their university department following academic mobbing and online academic bullying (Noakes & Noakes, 2021). At a large scale, othering entails a few dissidents on social media platforms being targeted for hypercriticism by gangstalkers.

Cyber gangstalking is a process of cyber harassment that follows cyberstalking, whereby a group of people target an individual online to harass him or her. Such attacks can involve gossip, teasing and bad-jacketing, repeated intimidation and threats, plus other fear-inducing behaviours. Skeptics' critical contributions can become swamped by pre-bunkers and fellow status-quo defenders. Such pseudo-skeptics may be sponsored to trivialise dissenters' critiques, thereby contributing to a fact choke against unorthodox opinions.

In Dr Christine Cotton's case in March 2022 her name was disclosed in a list as part of a French Senate investigation into adverse vaccine events. A ‘veritable horde of trolls seemingly emerged out of nowhere and started attacking’ her ‘relentlessly’ (p143). These trolls were inter-connected through subscribing to each others’ accounts, which allowed them to synchronise their attacks. They attempted to propagate as much negative information on Dr Cotton as possible in a ‘Twitter harassment scene’. Emboldened by their anonymity, the self-proclaimed “immense scientists” with masters in virology, vaccines, clinical research and biostatistics, launched a character assassination. They attacked her credentials and work history, whilst creating false associations (“Freemasonry” and “Illuminati”).

This suggests how identity politics sensibilities and slurs are readily misused against renegades. In the US, those questioning COVID-19 policies were labelled “far right” or “fascist”, despite promoting a libertarian critique of healthcare authoritarianism! In addition, orchestrators of cybermobbing tagged dissidents accounts to be those of someone who is: 'anti-science', 'an anti-vaxxer', 'biased', 'charlatan', 'celebrity scientist', 'conspiracy theorist', 'controversial', 'COVID-19 denier', 'disgraced scientist', 'formerly-respected', 'fringe expert', 'grifter', 'narcissist with a Galileo complex', 'pseudo-scientist', 'quack', 'salesman', 'sell-out' and 'virus', amongst other perjoratives. Such terms are used as a pre-emptive cognitive vaccine whose hypnotic language patterns ("conspiracy theorist") are intended to thwart audience engagement with critical perspectives. Likewise, these repeatedly used terms help grow a digital pillory that becomes foregrounded in the pattern of automated suggestions in search engine results.

In this Council of the Cancelled, Mike Benz, Prof Jay Bhattacharya, Nicole Shanahan and Dr Eric Weinstein speculate about hidden censorship architectures. One example is Google's automated tagging for "controversial" public figures. These can automatically feature in major mainstream news articles featuring COVID-19 dissidents. This is not merely a visual tag, but a cognitive tag. It marks "controversial" individuals with a contemporary (digital) scarlet letter.

In Dr Cotton's case, some trolls smeared her work in raising awareness of associations for the vaccine injured to be helping “anti-vaccine conspiracy sites”. She shares many cases of these injuries in her book and was amazed at the lack of empathy that Twitter users showed not just her, but also those suffering debilitating injuries. In response she featured screenshots of select insults on her blog at https://christinecotton.com/critics and blocked ‘hundreds of accounts’ online. In checking the Twitter profiles attacking her, she noticed that many with ‘behavioural issues were closeby’. Dr Cotton hired a ‘body and mind’ guard from a security company for 24-hour protection. Her account was reported for “homophobia”, which led to its temporary closing. After enduring several months of cyber-harassment by groups, a behaviour that can be severely be punished by EU law, Dr Cotton decided to file complaints against some of them. Christine crowdfunded filing legal complaints against Twitter harassers from a wide variety of countries. This complaint sought to work around how cyber harassers think anonymity is suitable for avoiding lawsuits for defamation, harassment and public insults.

#4 Not blocking impersonators or preventing brandjacked accounts

Impersonator's accounts claiming to belong to dissidents can quickly pop up on social media platforms. While a few may be genuine parodies, others serve identity jacking purposes. These may serve criminal purposes, in which scammers use fake celebrity endorsements to phish "customers" financial details for fraud. Alternately, intelligence services may use brandjacking for covert character assassination smears against dissidents.

The independent investigative journalist, Whitney Webb, has tweeted about her ongoing YouTube experience of having her channel's content buried under a fact choke of short videos created by other accounts:

Instead of blanket censorship, I am having YouTube bury all my actual interviews/content with videos that use short, out of context clips from interviews to promote things I would never and have never said. Below is what happens when you search my name on YouTube, every single… pic.twitter.com/xNGrfMMq52
— Whitney Webb (@_whitneywebb) August 12, 2024

Whether such activities are from intelligence services or cybercriminals, they are very hard for dissidents and/or their representatives to respond effectively against. Popular social media companies (notably META, X and TikTok) seldom respond quickly to scams, or to the digital "repersoning" discussed in a Corbett Report discussion between James Corbett and Whitney Webb.

In Corbett's case, after his account was scrubbed from YouTube, many accounts featuring his identity started cropping up there. In Webb's case, she does not have a public profile outside of X, but these were created featuring her identity on Facebook and YouTube. "Her" channels clipped old interviews she did and edited them into documentaries on material Whitney has never publicly spoken about, such as Bitcoin and CERN. They also misrepresented her views on the transnational power structure behind the COVID-19 event, suggesting she held just Emmanuel Macron and Klaus Schwab responsible for driving it. They used AI thumbnails of her, and superimposed her own words in the interviews. Such content proved popular and became widely reshared via legitimate accounts, pointing to the difficulty of dissidents countering it. She could not get Facebook to take down the accounts, without supplying a government-issued ID to verify her own identity.

Digital platforms may be disinterested in offering genuine support- they may not take any corrective action when following proxy orders from the US Department of State (aka 'jawboning') or members of the Five Eyes (FVEY) intelligence agency. In stark contrast to marginalised dissenters, VIPS in multinationals enjoy access to online threat protection services (such as ZeroFox) for executives that cover brandjacking and over 100 other cybercriminal use-cases.

#5 Filtering an account's visibility through ghostbanning

As the Google Leaks (2019) and Facebook- (2021) and Twitter Files (2022) revelations have spotlighted, social media platforms have numerous algorithmic censorship options, such as the filtering the visibility of users' accounts. Targeted users may be isolated and throttled for breaking "community standards" or government censorship rules. During the COVID-19 event, dissenters' accounts were placed in silos, de-boosted, and also subject to reply de-boosting. Contrarians' accounts were subject to ghostbanning (AKA shadow-banning)- this practice will reduce an account’s visibility or reach secretly, without explicitly notifying its owner. Ghostbanning limits who can see the posts, comments, or interactions. This includes muting replies and excluding targeted accounts' results under trends, hashtags, searches and in followers’ feeds (except where users seek a filtered account's profile directly). Such suppression effectively silences a user's digital voice, whilst he or she continues to post under the illusion of normal activity. Ghostbanning is thus a "stealth censorship" tactic linked to content moderation agendas.

This term gained prominence with the example of the Great Barrington Declaration's authors, Professors Jay Bhattacharya, Martin Kulldorff, and Sunetra Gupta. Published on October 4, 2020, this public statement and proposal flagged grave concerns about the damaging physical and mental health impacts of the dominant COVID-19 policies. It argued that an approach for focused protection should rather be followed than blanket lockdowns, and that allowing controlled spread among low-risk groups would eventually result in herd immunity. Ten days later, a counter- John Snow Memorandum was published in defence of the official COVID-19 narrative's policies. Mainstream media and health authorities amplified it, as did social media given the memorandum's alignment with prevailing platform policies against "misinformation" circa-2020. In contrast, the Great Barrington Declaration was targeted indirectly through platform actions against its proponents and related content:

Stanford Professor of Medicine, Dr Jay Bhattacharya’s Twitter account was revealed (via the 2022 Twitter Files) to have been blacklisted, reducing its visibility. His tweets questioning lockdown efficacy and vaccine mandates were subject to algorithmic suppression. Algorithms could flag his offending content with terms like “Visibility Filtering” (VF) or “Do Not Amplify”, reducing its visibility. For instance, Bhattacharya reported that his tweets about the Declaration and seroprevalence studies (showing wider COVID-19 spread than official numbers suggested) were throttled. Journalist Matt Taibbi's reporting on the "Twitter Files" leaks confirmed that Twitter had blacklisted Prof Bhattacharya's account, limiting its reach due to his contrarian stance. YouTube also removed videos in which he featured, such as interviews in which he criticised lockdown policies.

The epidemiologist and biostatistician, Prof Kulldorff observed that social media censorship stifled opportunities for scientific debate. He experienced direct censorship on multiple platforms, which included shadowbans. Twitter temporarily suspended his account in 2021 for tweeting that not everyone needed the COVID-19 vaccine ('Those with prior natural infection do not need it. Nor children'). Posts on X and web reports indicate Kulldorff was shadowbanned beyond this month-long suspension. The Twitter Files, released in 2022, revealed he was blacklisted, meaning his tweets’ visibility was algorithmically reduced. Twitter suppressed Kulldorff's accurate genetic vaccine critique, preventing comments and likes. Internal Twitter flags like “Trends Blacklisted” or “Search Blacklisted” (leaked during the 2020 Twitter hack) suggest Kulldorff's account was throttled in searches and trends, a hallmark of shadowbanning where reach is curtailed without notification. Algorithmic deamplification excluded Prof Kulldorff's tweets from being seen under trends, search results, or followers’ feeds- except where users sought his profile directly. This reflects how social media companies may apply visibility filters (such as a Not Safe For Work (NSFW) view). Kulldorff also flagged that LinkedIn’s censorship pushed him to platforms like Gab, implying a chilling effect on his professional network presence.

An Oxford University epidemiologist, Professor Gupta faced less overt account-level censorship, but still had to negotiate content suppression. Her interviews and posts on Twitter advocating for herd immunity via natural infection amongst the young and healthy were often flagged, or down-ranked.

#6 Penalising accounts that share COVID-19 "misinformation"

In addition to ghostbanning, social media platforms could target accounts for sharing content on COVID-19 that contradicted guidance from the Global Private Partnership (GP3)'s macro-level stakeholders, such as the Centre for Disease Control or the World Health Organisation. In Twitter's case, it introduced a specific COVID-19 misinformation policy in March, 2020, which prohibited claims about transmission, treatments, vaccines, or public health measures that the COVID-19 hegemony deemed “false or misleading.” Such content either had warning labels added to it, or was automatically deleted:

Tweets with suspected MDM were tagged with warnings like “This claim about COVID-19 is disputed” or with labels linking to curated "fact-checks" on G3P health authority pages. This was intended to reduce a tweet’s credibility without immediate removal, whilst also diminishing its poster's integrity.

Tweets that broke this policy were deleted outright after flagging by automated systems or human moderators. For instance, Alex Berenson’s tweets questioning lockdown efficacy were removed, contributing to his eventual ban in August 2021. In Dr Christine Cotton's case, Twitter classified her account as “sensitive content”. It gradually lost visibility with the tens of thousands of followers it had attracted. In response, she created a new account to begin ‘from scratch’ in August 2022. The Twitter Files revealed that such censorship was linked to United States government requests (notably from the Joe Biden administration and Federal Bureau of Investigations). For example, 250,000 tweets flagged by Stanford’s Virality Project in 2021 were removed by Twitter.

In March 2020, Meta expanded its misinformation policies to target COVID-19-related MDM. Facebook and Instagram applied content labelling and down-ranking, with posts allegedly featuring MDM being labeled with warnings (such as 'False Information' or 'See why health experts say this is wrong') that linked to official sources. Such posts were also down-ranked in the News Feed to reduce their visibility. Users were notified of violations and warned that continued sharing could further limit reach or lead to harsher action. In late 2021, down-ranking also became applied to “vaccine-skeptical” content not explicitly violating rules but potentially discouraging vaccination. Posts violating policies were removed outright.

With LinkedIn's smaller, professional user base, and the platform's lower emphasis on real-time virality, led it to prefer the outright removal of accounts over throttling via shadow-bans. Accounts identified as posting MDM could face temporary limits, such as restricted posting privileges or inability to share articles for a set period. LinkedIn users received warnings after a violation, often with a chance to delete the offending post themselves to avoid further action. Such notices cited the policy breach, linking to LinkedIn’s stance on official health sources. This approach to COVID-19 MDM followed LinkedIn’s broader moderation tactics for policy violations.

In Dr Cotton's case, she shared her Pfizer COVID-19 clinical trial's critique on LinkedIn to get feedback from her professional network of former contacts in the pharmaceutical industry. This first post was removed within 24 hours (p.142), and her second within an hour. This hampered her ability to have a debate on the methodology of Pfizer's trial with competent people. Prof Kulldorff also had two posts deleted in August 2021: one linking to an interview on vaccine mandate risks and another reposting Icelandic health official comments on herd immunity.

Accounts that posted contents with links to external, alternate, independent media (such as Substack articles or videos on Rumble) also saw such posts down-ranked, hidden or automatically removed.

This is the first post on techniques for suppressing health experts' social media accounts (and the second on COVID-19 censorship in the Fifth Estate). My next in the series will address more extreme measures against COVID-19 dissidents, with salient examples.

Please follow this blog or me on social media to be alerted of the next post. If you'd like to comment, please share your views below, ta.

Wednesday, 5 February 2025

Celebrities cannot stop their brandjacking, since many authorities are unable to help!

Written for cyber- and digital crime researchers and reporters. Plus, brandjacked celebrities and their representatives.

In discussion with reporters and PR experts, most seemed to be under the mistaken impression that celebrities enjoy a viable route to stop their brandjacking in scam adverts on social media platforms. I wrote this post to explain that although SA public figures seem well-resourced and influential, none have a viable route to prosecuting their brandjackers due to an absence of support from many authorities:

Since 2019, The Noakes Foundation has supported research into the brandjacking of influential celebrities' reputations on social media, and other poorly moderated platforms. The Fake Celebrity Endorsement (FCE) research team is documenting how this digital crime is an inscrutable, traumatic experience for celebrities, their representatives, and the financial victims who report being conned by fake endorsements. In addition to being traumatised by being featured in fake adverts, microcelebrities are further traumatised by the many reports from fans of their upset at being conned. A few celebrities have become targets for recurring cybervictimisation with no recourse, resulting in repeat trauma.

The FCE distinguishes 'digital crimes' from 'cybercrimes': Micro-fraudsters typically target private individuals with limited access to resources for combating digital crime. This contrasts to cybercrimes in which corporations are attacked (Olson, 2024). They are often well positioned to support their employees with costly resources that private individuals cannot afford. Research into the latter is well-resourced, as is interventions to stop it. By contrast, the fighting of digital crimes that impact private citizens are poorly resourced, particularly in the Global South. In the case of fake celebrity endorsements, press reports of this scam suggest that the problem grows each year- eleven South African celebrities fell victim to it in 2024, up from two in the first reports of 2014.

Fake celebrity endorsement is a digital crime that may require many authorities in society to combat it. Below is a list of the roleplayers that might potentially help prevent digital crimes:

1) celebrity influencers,

2) financial victims,

3) social media advertising providers,

4) Poorly-moderated content hosts,

5) banks,

6) cyber defence companies,

7) cybercrime reporters and statistics gatherers (industry researchers),

8) cybercrime educators,

9) anti-crime activists (PBOs and NGOs),

10) social media platforms (eg Big Tech),

11) financial investors,

12) government politicians,

13) the police,

14) international law enforcement,

15) local law,

16) Higher Education and its funders,

17) Product regulators.

While digital crime victims might expect support from these thirteen other role-players, this post spotlights their limitations. Some simply are unable to prioritise fighting fake celebrity endorsements, while others' interests may not be served in tackling this crime!

Figure 1 - the brandjacking digital crime process

Figure 1 shows a simplified process of the fake endorsement phishing scam. The authors of this digital crime are unknown- they can range from gangs, to the invisible threat of AI and bot armies, to even military intelligence agencies raising funds. Not only do these cybercriminals exploit scamming ecosystems inside popular social media platforms, they also exploit related ecosystems on platforms such as Huione Guarantee (now "Haowang Guarantee"), a Cambodian conglomerate. It offers a messaging app, stable coin, and crypto exchange, and has facilitated $2 billion in transactions. Such platforms are integral to the industrialisation and scaling-up of online scams, for example through supporting the outsourcing of scammers' money-laundering activities (The Economist, 2025).

1) Celebrity influencers

On digital media, celebrity influencers are 'micro-celebrities', who can also be 'influencers' (if paid to share content). Micro-celebrities may not be aware of the dangers of hyper-visibility, since there are no 'Here Be Dragons' signs at the on-ramps to creating their digital profiles. Here, celebrities agree to legal contracts that are heavily one-sided in favour of social media platforms versus users (Sarafian, 2023). These contracts do not place an onus on social networks to warn or protect their users from digital visibility risks; such as brandjacking and impersonation. The FCE project has approached almost 50 South African celebrities via their agents to participate in its research. Each's reputation was reported to have been stolen for scam adverts. Despite offering incentives, only three (plus select representatives) agreed to participate. Most may want to put their negative experiences behind them, while fearing reputational risks from being involved in a research process they are unfamiliar with, and whose outputs may be misperceived as potentially damaging. So, a big challenge exists in persuading micro-celebrities to contribute their experiences to research, so that they can be shared to inform digital crime fighters' responses.

2) Financial victims

Fans who have developed a parasocial relationship with a particular celebrity they follow, may genuinely believe that the fake endorsement adverts are a legitimate offer. Notwithstanding, the product’s promise seeming to be too good to be true. Having been conned, its victims may be ashamed, or in denial. Many may consider their financial loss not worth reporting (as a micro-fraud versus a serious crime). Even if victims are willing to report the digital crime, it may not be obvious which authority the crime is best reported to.

3) Social media advertising services

Online advertisers, and digital platforms, may not understand or monitor the threat of digital crimes, such as celebrity brandjacking. It is not well-defined and may also be challenging to report on, since it spans several crimes itself: 1) impersonation; 2) non-consensual image sharing; and 3) the infringement of a public figure's intellectual property (through copyright violation of still images and audio-video). In addition, the crime causes; 4) reputational damage by suggesting a public figure’s association with a scam that often involves 5) financial fraud and hacking. Social media advertising complaint reporting formats only permit the reporting of one type of infringement at a time. This potentially leaves a blindspot, as users cannot report all the aspects that characterise the celebrity brandjacking crime. If it is a widespread problem, social media advertisers may also prefer not to flag it as a concern, thereby protecting their public reputations. Albeit, at the expense of celebrity and other financial victims.

4) Poorly-moderated content hosts

To make their offers seem more credible, scammers also post fake content on poorly-moderated sites (such as "clickbait news", "positive reviews" on online forums, and "scientific papers" on academic social networks). Even if such fake content is reported and removed, it can be quickly shifted by scammers to worse-moderated hosts...

5) Banks

As the financial victims legitimately authorise payments off their own accounts, victims do no enjoy recourse via their banks. To avoid new transactions from scammers, these victims often have to pay banks for new cards after terminating their old ones. It is unclear what role banks could adopt in combating digital crimes, wherein clients are defrauded whilst following a seemingly legitimate payment process.

Figure 2 - Authorities who could contribute to fighting digital crimes

6) Cyber defence companies

Cyber defence businesses are focused on providing profitable services to corporates. Such services are often unaffordable to even the wealthiest celebrities in SA. However, some may be fortunate to work for companies that use cyber defence services that pro-actively monitor cyberspace, and warn employees against digital impersonation and related risks. Such services include Darkivore, Flashpoint, Netcraft (FraudWatch International), SGS and ZeroFox. It does not seem that cyber defence companies can produce a profitable service that supports rapid responses to fake endorsements and related crimes. Even if it such a service was not unaffordable, most SA celebrities have not been targeted for revictimization. So it seems unlikely they would subscribe to such a service annually.

7) Cybercrime reporters and statistics gatherers

While there have been many reports for weight-loss and money-making cryptocurrency scams featuring particular celebrities, the media, celebrities agents and PR companies seem to report on these brandjackings as once-off events. Reports typically cover the latest flare-up to negatively impact one or two stars, plus their fans. At the same time, cybercrime statistics do not include this digital crime, whose costs to victims in SA are unknown, and difficult to aggregate. This points to a need for developing an holistic view of digital crime from consolidated reports. Research into digital crimes that can bridge the work of journalists and crime statisticians seems urgently needed to describe such crime's extant, frequency and costs to society. Developing robust reporting mechanisms for digital crimes (particularly ones that are challenging like 'fake social media adverts for phishing' {as they include several sub-crimes}) would seem an important contribution that law enforcement, researchers and statisticians can make. Reporters and researchers can also develop robust definitions of emergent digital crimes to grow awareness of them. This should aid more accurate reports.

8) Cybercrime researchers, and educators, in companies
In a similar collaborative vein, cybercrime researchers and educators in companies are working together to help combat digital crimes targeting their employees and clients. In particular, banks and insurance companies in SA are pro-actively raising awareness around phishing and other common digital crimes. This is done in communications that range from email newsletters, to pop-up warnings that clients must acknowledge reading post log-in.

9) Anti-crime activists (PBOs and NGOs)
Anti-digital crime education tends to focus on educating high school students and working professionals with preventative knowledge in English. However, our research into fake celebrity endorsement victims' online commentary suggest that most are over fifty five, with English being their second language, at best. In response, The Noakes Foundation has supported the development of modules in English for educating silver surfers on the most common digital crimes. Ideally, though, these modules (and reportfakeendorsement.com's content) should be available in South Africa's 11 official languages.

10) Social media platforms and their Big Tech owners
Social media companies, and their Big Tech owners, would seem to have a particular responsibility for protecting users from digital crime threats on their platforms. In contrast, there is a decade-long history in SA of even influential celebrities not being well-supported via speedy responses to their brandjacking, and scam adverts are seldom taken down based on celebrities', their representatives' and other victims' reports.

The most popular platform for this scam in SA is Meta's Facebook and Instagram. Meta does not understand the content that its users share (Horwitz, 2023). Further, it does not report on scam ecosystems based inside its own platforms. Consequently, neither Facebook nor Instagram can pro-actively identify digital crimes, let alone quickly adapt their systems to stop emergent threats from micro-fraudsters. It's left up to whistleblowers, former employees, investigative journalists and researchers to create awareness on these platforms' serious flaws, such as it being used as a scammers' ecosystem tied to scam-as-service ones. This would seems at odds with corporate responsibility- Meta should publicly report regarding its progress in tackling scam-ecosystems on its FB, WhatsApp and Instagram platforms. It could also pro-actively warn vulnerable users, such as the aged, against the latest scam risks.

In a sense, digital crimes by cybercriminals on social networks can be considered a parasitic attack within a larger parasitic host: Meta’s Facebook and Instagram are infomediaries that misrepresent themselves as symbionts in supporting users’ communal connections online. In reality, Meta’s business model is parasitic in relying on 3 000 000 000 users to share content (Robbins, 2018). Much of this content is not the work of original/creative producers, but rather sampled from content that's proved popular on other platforms. In essence, social media platforms are middlemen between content creators and their audiences, taking most of the profits from advertising. These platforms also take the intellectual property of online content creators. In the Global South this serves as a form of neocolonial data extraction as Big Tech multinationals from the Global North extract its data, with little being reciprocated. For example, while powerful celebrities in the US can enjoy access to dedicated Facebook support, there is no equivalent offering for influential SA users. Instead, they are lucky to stumble onto internal staff or Trusted Partners who can best help them respond to the Facebookjacking or Instajacking crimes.

In contrast to the usefulness of human insiders, reports to Meta's AI that manages users' reports of dubious accounts and content is simply not capable of recognising malicious advertisers' accounts; At face value, there is in nothing “wrong” with how the scammers’ accounts are set up - They have have a human profile (fake name) managing a business profile (fake name and business). Reporting the scam accounts is useless, since the fraudsters fill-in all the right criteria to fly under the radar! The scammers use 'Like Farms' and a network of fake profiles to all create a sense of legitimacy through 'liking, sharing and commenting' on posts and ads. The criminals also use a “legitimate website” - this is a bought domain and hosing and (questionable design) - selling a “product” and accumulating data of visitors' info and credit card details. All this seems to be legitimate business behaviour to AI, but is malicious and AI cannot detect that. Scammers use a (stolen) credit card, or hijacked meta Ads manager profile and run advertIf this content had truly been verified by a human it would have been taken down immediately. Even to the most untrained eye it was obvious that this content was a deep fake.s through their “business page”. This works for a short while until the card or the account are stopped, and then they just create another one. These ADs are selling a product online, the product is seemly harmless and well within the legal parameters of Meta's Community standards. The fact that its a fake product is immaterial to Meta, the onus in on the customer to know when they are being scammed, and if users try to report this to be a harmful product, it doesn’t work” as that is deemed a matter of personal opinion! Where such content is checked by a human moderators, the content is so obviously fake that they taken it down quickly.

It appears that Meta's Facebook and Instagram are turning a blind eye to this digital advertising crime. The benefit to META is clear with them reaping the rewards through advertisers' spending: Trustfull's 2024 report expects Deep Fake Fraud to reach $ 15.7 billion in 2024. Meta is set to take a large chunk of that ad-spend revenue in distributing fake, malicious content. It’s hard not to draw to the conclusion that it seems irrelevant to META if the content is genuine or a scam, or if the account used to promote these scams has been hacked or cloned. Either way, META and Facebook still profits.

11) Financial investors

Investors are focused on the bottom-line of financial profit. To achieve it, social media platforms' developers spotlight the metrics of an ever-increasing flow of communication marking their platforms' commercial expansion. Given this all-consuming quantitative focus, it's unsurprising that these platforms' developers and investors are largely disinterested in paying the costs to understand the negative experiences on platforms. Particularly when combating these might impede their growth and monetisation!

12) Government politicians

SA's parasitic political class has been slow to take action for protecting its citizenry from the excesses of social media platforms, and digital crimes on them. For example, it has not protected the intellectual rights of original digital content producers by passing a Creators Bill of Rights to limit their online exploitation. On social media, SA creatives do not retain copyright, have rights of termination and appeal, etc. Globally, most online content creators struggle to make a living from the work that they do. Social media companies' oligarchic power and their poor regulation by the law contributes to this. Further, policy inactivity may suggest that government decision makers' are guided by Big Tech's funding and political support, more-so than its digital creatives' rights and needs.

13) Local police

Brandjacking is not viewed as a serious crime by SA authorities who might be expected to intervene as guardians. Cybercrime experts in the SA police have to focus their limited resources on fighting severe digital crimes (like the online trafficking of children, drugs, and guns). They simply cannot address digital crimes that have not been shown to have serious impacts on their victims, such as phishing micro-frauds.

14) International law enforcement

In contrast to under-resourced local law enforcement authorities, global ones (such as Interpol) are better resourced to potentially offer some form of response to digital crimes on social media. However, until decent stats and reports for digital micro-frauds are documented and shared with global authorities, these digital crimes are not notifiable, so cannot be directly investigated at an international level.

15) Local law

Even if a foreign criminal network behind a scam is found though investigation, and the legal frameworks exist for their extradition, the costs for local law enforcement to prosecute scammers may well prove prohibitive to the State. The global proceeds of online fraud are probably more than $500bn a year, so another major concern is that online fraudsters have become rich and powerful enough to corrupt entire governments (Scam Inc podcasts, 2025). Scam "businesses" can turn countries into the cyber-scam equivalent of narco-states, and their operations can be found all over the world. Broadly, 1.5 million scammers are at work, from Namibia up to the Isle of Man, and from Mexico across to Fiji (The Economist, 2025). Where scam bosses have strong clout within a political system, it becomes impossible to enact policies that undermine their fraud. Corrupted states, such as Cambodia, would seem unlikely to extradite criminals who pose a reputational risk in potentially implicating senior state officials. Extradition from lawless places, such as Myanmar, is also impossible.

16) Higher Education and research funders

Like banks, universities are stepping up their preventative digital crime awareness communications, and attracting research funding to build scholarship into cyber- and digital crimes. More generally, universities can lead discourse on the digital crimes issue, catalysing inter- and trans-disciplinary collaborations. Funders of university grants might support design thinking, or strategic design activities to develop solutions for the seemingly intractable brandjacking micro-fraud. Likewise, ethical research into the issues that emerge in researching digital crimes under fake personas. Given that the brandjacking of influential scholars would seem to also pose reputational risks to their university-as-employer, related research could be motivated as a neglected, but potentially valuable contribution.

17) Product regulators

Fake celebrity endorsements typically promote dubious products, that may actually be delivered. As such, customers may assume that they are protected by local regulators for those particular product types. For example, British doctors have been brandjacked to market "wonder drugs" that cure blood pressure. "Their" customers might assume that they are protected by the UK's General Medical Council. However, this falls outside the GMCs remit, which can only address with promotions from genuine doctors on its register. The GMC cannot tackle 'computer generated videos' by unknown fraudsters (Stokel-Walker, 2014). It seems unlikely that any product regulator can help with tackling fake products marketed by unregistered and anonymous cybercriminals.

Celebrity, you are on your own in responding to digital crime?!

It should be clear that celebrities are unlikely to be supported in stopping fake social media adverts. While five key authorities are working to raise awareness against digital crimes (5, 7, 8, 9 & 16), there is little-to-no help available from eight (3, 4, 6, 12, 13, 14, 15 & 17). At the digital crime's fountainhead (10 & 11), social media platforms are actually disincentivised against tackling the problem. In the absence of support from criminal prosecutors, the law or cybercrime fighting businesses, it should be clear that the guardianship role of social media companies is non-negotiable. Social media platforms that are heavily used by scammers, could take responsibility for this by entrusting micro teams of moderators around the world to review adverts. Verification by well-trained human resources who can disable fraudulent accounts seems to be the best answer to stopping brandjacking on social media. The current AI approach is failing miserably, with counter-technologies many steps behind cybercriminals' "innovations".

There is not enough being done at an entry level to assist smaller companies and the general public who are constantly being attacked by digital crimes. The accounts of celebrities, plus their representatives, whom we've interviewed suggest that pressure must urgently be placed on social media platforms to provide effective brandjacking reporting, and prevention, tools. Without these, deep fake adverts can spread quickly- reaching tens-of-thousands of people a day. At the same time, celebrities must focus on building rapid awareness in the media, plus at the site of the digital crimes, to alert potential victims.

Just as celebrity co-operation is important, so is that by civil society organisations who can collaborate for pressuring authorities to act in a more responsible and pro-active manner in tackling digital crime. For example, Anna Collard (SVP Content Strategy and Evangelist at KnowBe4 Africa) is doing important work in building networks that can collaborate in educating the public around the dangers of cyber- and digital crimes. Likewise, international and local networks must motivate for sounder strategic interventions from key role players to thwart global networks.

Serious funding is also needed to support awareness programs for educating vulnerable groups. In particular, communities like the elderly are a very high risk, though being less tech- and media savvy. Public benefactors are also needed to assist educational initiatives, like ReportFakeEndorsement, with reaching a broad audience and for greatly increasing the research being done into new digital crimes, and how to thwart them.

Please comment with suggestions to improve this post

I am not an expert on all 13 authorities and their responses, so welcome any constructive corrections for improving what's been shared above. Please comment below, so that I can review your feedback, and perhaps update this blogpost, acknowledging you for your advice, below.

Acknowledgements

Thanks to Dr Taryn van Niekerk for proposing the term ‘unchartered territory’ to describe celebrities’ (and their reps) challenges as novices responding to a brandjacking. That insight helped inspire this post, which maps authorities in an 'unknown territory', and the support they might offer, or cannot. Also, The Noakes Foundation and the FCE team appreciate Mr Byron Davel's advice regarding the ZeroFox offering. Plus the broader field of brand protection, and cyber-defence against dark web operations. TNF also appreciates the journalist Lyse Comins' critique of Meta's slow action in her news article- Meta criticised for slow action as deepfake adverts target South African celebrities.

Thursday, 29 August 2024

After half-a-million views, "Dr Noakes" erection dysfunction "advert" taken down by Facebook + suggested actions for META to do better

I am pleased to report that The Noakes Foundation has succeeded in getting a fake 'Dr Noakes' advert for erectile dysfunction pills removed from META. This is after a month of trying varied methods without success to stop the brandjacking of Professor Tim Noakes' identity, and his impersonation via deepfake reels and accounts on Facebook.

Brandjacking is the ‘allegedly illegal use of trademarked brand names - on social network sites’ (Ramsey, 2010 p851). Cybercriminals misuse the trademarks of others without authorization. For example, ‘Facebookjacking’ and ‘Instajacking’ see public figures’ usernames, account names, and/or digital content being used for fake accounts and video "adverts" on Meta’s respective popular social networks- Facebook and Instagram. Such brandjacking via fake celebrity endorsement spans several types of crime: (1) Impersonation, (2) Non-consensual image sharing, and (3) the Infringement of a public figure's intellectual property through copyright violation of still images and audio-video. In addition to causing (4) Reputation damage to the public figure through suggesting association with a scam, cybercriminals may use it for (5) Financial fraud and hacking. Given that these are serious crimes, it is worrying that public figures in South Africa seem to receive minimal, if any, support from social media companies for stopping the fake endorsement digital crime. There is also a gap in scholarship for how public figures worldwide, and in SA, might best tackle this persistent crime.

Figure 1. Screenshot from the fake 'Dr Noakes' erection dysfunction advert on Facebook (2024)

On Thursday the 25th of July we were first alerted to a deep fake advert featuring Emeritus Professor Tim Noakes that ran on META's Facebook, and Tik Tok. As Figure 1 shows, the Facebook advert had been viewed over 584,000 times, liked by 637 accounts, and received 56 comments. While much of the likes and comments may be from bots, such high viewership of the reel itself is highly concerning. It suggests how rapidly a cybercriminals' adverts spread to potential victims- at over 16,000 views per day!

Figure 2. Screenshot of scammers' Facebook account featuring "Dr Tim Noakes" erection pill adverts (2024)

Figure 3. Scammer account location behind fake Facebook Dr Tim Noakes adverts (2024)

Our initial Facebook advert lookup revealed that one page was running four adverts (Figure 2). This account ("Tristan") was managed from Nepal and India (Figure 3).

Figure 4. Screenshot of fake Tristan account header behind Dr Tim Noakes adverts on Facebook.jpg

Figure 4. Screenshot of fake Tristan account header behind Dr Tim Noakes adverts on Facebook (2024)

This fake account page also leveraged fake interactions to suggest that it was liked, and followed (Figures 4 and 5).

Figure 5. Screenshot of fake Tristan account header behind Dr Tim Noakes adverts on Facebook

Figure 5. Screenshot of fake "Tristan" Facebook account details behind Dr Tim Noakes adverts (2024)

This account was reported to Facebook via a third-party. During this “warning period”, the account's owners launched four new "Dr Tim Noakes" campaigns. Each was documented and reported to Facebook. Interestingly, the links to the online store “sites” were dead ends. However, a 'Call Now' button could still support a call agent's phishing of victims financial details.

The absence of a link for data gathering suggested that this scam was primarily not for phishing such sensitive data, or selling fake products. Rather the advert's design seems geared for stealing advertising revenue via deepfake creation. The scammers hack into the "advertiser"'s Meta account to distribute fake adverts that run up tens-of-thousands of dollars in spend. In this case it was a government-based account from an unknown location. Such adverts may also carry malware, with users clicking on them being vulnerable to hacking. These paid ads also have the impact of pushing potential followers to the advertiser’s page. More followers results in more people seeing the content, and Meta indirectly benefiting from the cybercrime's increased visibility by achieving higher advertising rates.

Figure 6. Screenshot of scammers' "Hughles" Facebook account (2024)

Figure 7. Screenshot of scammers' "Hughles" Facebook account's Dr Tim Noakes adverts (2024)

The scammers flick-flacked between varied accounts in committing this cybercrime- they initially used "Hughles" (Figures 6 and 7), "Cameron Sullivan Setting", and "Murthyrius" in launching the same deepfake ads. By the 28th of July, 13 of these "adverts were taken down by Facebook, but the scammers shifted to new accounts, "Longjiaren.com" (Figure 8) and "Brentlinger" (renamed "Brentlingerkk" after we reported it). On the 29th of August, these accounts and their adverts were disabled by Facebook.

Screenshot of Longjiaren.com scammers Facebook account for fake adverts.jpeg

Figure 8. Screenshot of Longjiaren.com scammers Facebook account for fake adverts (2024)

Such adverts typically reach viewers outside The Noakes Foundation, Nutrition Network and Eat Better South Africa’s networks. Their audiences know Professor Noakes does not endorse miracle weight loss and other cures. To reach vulnerable publics, The Noakes Foundation has run Facebook alerts to warn about this latest cybercrime. Ironically, the most recent advert attempting to flag the "Dr Noakes" scam was blocked by Facebook advertising (Figure 9)!

Screenshot of Facebook rejecting anti scam ad from The Noakes Foundation.jpg

Figure 9. Facebook rejects anti scam ad from The Noakes Foundation (2024)

Actions for META to do better in fighting cybercrime on its platforms

As Anna Collard (KnowBe4) spotlights in her recent interview with eNews, social media platforms are a vital source for news in Africa. Consequently, these platforms must be held more accountable for any slow responses to synthetic- and deep-fakes. It is greatly concerning that META's Facebook platform is so rife with many serious crimes (ranging from sextortion and child-trafficking to drug pushing).

META can be more pro-active in tackling such cybercrimes {plus less serious ones like fake celebrity endorsement}, by prioritising these seven steps below:

1) Actively communicate that all users' must have a 'zero trust' mindset;
2) Create a compliance team that is dedicated to thwarting cybercriminals' activities;
3) Offer at least one human contact on each META platform for serious reports of criminal misuse;
4) Promote frequent reporters of cybercrime by referring them to META's Trusted Partners or Business Partners for rapid aid;
5) Encourage external research on every platform regarding cybercriminals' activities (such initiatives could develop inexpensive tools. For example, for celebrities' reps to protect public figures from being deep faked in "adverts");
6) Provide more feedback on what was influential in reporting cybercrime for accounts and content to be removed. Without such feedback, fraud reporters may not be sure which reports are most effective;
7) Have a recommendation system in place for support networks that cybervictims can approach (such as referring South Africans to its national CyberSecurity hub).

In addition, META might consider these suggestions from The Noakes Foundation's Report Fake Endorsement initiative, to: (8) enhance deepfake detection technology, (9) apply stricter verification processes, (10) increase transparency and reporting tools, (11) support local educational initiatives, (12) promote collaborations with local cybercrime experts, (13) implement proactive monitoring systems to detect unusual patterns in ads, and (14) reinforced consequences for violations.

By sharing this "Dr Noakes" case study (and developing others), The Noakes Foundation hopes to raise awareness of the fake celebrity endorsement cybercrime, plus the importance of Big Tech guardians stepping up to fulfil their responsibilities. We are also liaising with sympathetic allies (KnowBe4® Africa Security Awareness, Orange Defence, Wolfpack Information Risk and others) to grow the networks necessary to better support cybercrime prevention in South Africa.

Much can be done for targeted digital literacy education for vulnerable targets of cybercrime (such as #StopTheScam for silver surfers). We will also continue advocating that capable guardians (such as META, Twitter and TikTok) become more pro-active in protecting vulnerable publics on their platforms. Their gatekeeping role is vital, as the traditional bulwarks against crime (education, the police and the law) seem unable to catch-up with the "evolution"of global cybercrimes!

Subscribe to: Posts ( Atom )