Fakes

The interim report of the Australian Senate's Select Committee on Adopting Artificial Intelligence feastures the following recommendations 

 Recommendation 1 The committee recommends that, ahead of the next federal election, the government implement voluntary codes relating to watermarking and credentialling of AI-generated content. 

Recommendation 2  The committee recommends that the Australian Government undertake a thorough review of potential regulatory responses to AI-generated political or electoral deepfake content, including mandatory codes applying to the developers of AI models and publishers including social media platforms, and prohibitions on the production or dissemination of political deepfake content during election periods, for legislative response prior to the election of the 49th Parliament of Australia. 

Recommendation 3  The committee recommends that laws restricting the production or dissemination of AI-generated political or electoral material be designed to complement rather than conflict with the mandatory guardrails for AI in high-risk settings, the recently introduced disinformation and misinformation reforms, and foreshadowed reforms to truth in political advertising. 

Recommendation 4  The committee recommends that the Australian Government ensure that the mandatory guardrails for AI in high-risk settings also apply to AI systems used in an electoral or political setting. 

Recommendation 5  The committee recommends that the government examine mechanisms, including education initiatives, to improve AI literacy for Australians, including parliamentarians and government agencies, to ensure Australians have the knowledge and skills needed to navigate the rapidly evolving AI landscape, particularly in an electoral context.

Quackery

'Vaccine Misinformation for Profit: Conspiratorial Wellness Influencers and the Monetization of Alternative Health' by Rachel E Moran, Anna L Swan and Taylor Agajanian in (2024) 18 International Journal of Communication 1202–1224 comments 

Influencers in the alternative health and wellness space have leveraged the affordances of social media to make posting misleading content and misinformation a lucrative endeavor. This research project extends knowledge of antivaccine misinformation through an examination of the role of social media influencers and the parasocial relationships they build with audiences in the spread of vaccine-opposed messaging and how this information is leveraged for profit. Through digital ethnography and media immersion, we focus on three prominent antivaccine influencers—the Wellness Homesteader, Conspiratorial Fashionista, and Evangelical Mother—analyzing how they build community on Instagram, promote antivaccination messaging, and weaponize this information to direct their followers to buy products and services. 

Misinformation is an immensely profitable endeavor. Amplifiers of misinformation have found routes to monetize their digital content by using it to direct their online followers to purchase the products and services they endorse. Far-right news and opinion site Infowars, for instance, made $165 million between 2015 and 2018, selling health supplements and merchandise through the Infowars store (Vaillancourt, 2022) advertised during Alex Jones’ talk radio shows, often attached to misinformation narratives or in the context of discussing conspiracy theories (Locker, 2017). This project explores how misinformation is monetized, focusing specifically on how influencers within the antivaccination movement use social media to amplify misleading information about vaccinations and leverage this information for profit. 

Although vaccine misinformation far predates COVID-19, its scale and prominence have increased immensely because of the pandemic (Wardle & Singerman, 2021). Extant research has identified a range of vaccine-related misinformation, including spurious claims that the vaccine contains microchips (Virality Project, 2022) and broader attacks on the safety, efficacy, and necessity of COVID-19 vaccines (Brennen, Simon, Howard, & Nielsen, 2020). Further research has explored the dominant sources of vaccine misinformation, identifying the spread of vaccine opposition from antivaccine influencers (Center for Countering Digital Hate [CCDH], 2021)—in addition to a top-down amplification of misinformation from political elites (Enders, Uscinski, Klofstad, & Stoler, 2020). 

Alternative health and wellness influencers were a cause for concern during the COVID-19 pandemic because of their ties to misinformation and vaccine hesitancy (Maloy & De Vynck, 2021). Leveraging a lack of trust in Western institutionalized medicine, some wellness influencers have pushed hyperindividualistic frameworks that dispute the need for collective vaccine uptake in favor of natural wellness (Kale, 2021). Furthermore, the sociotechnical savvy of wellness influencers affords them significant reach for their content. A report from the CCDH (2020) noted that the top 12 antivaccine influencers gained 877,000 followers between December and June 2020 (p. 5). Beyond numerical reach, the parasocial relationships built via social media exacerbate the impact of vaccine misinformation. Moreover, influencers well-versed in the economic and technical infrastructures of social media are well positioned to financially benefit from the misinformation they share. 

This article opens by discussing research on the spread of vaccine-related misinformation on social media and within the health and wellness space, as well as the role of parasocial relationships in this spread. By highlighting the role of gender in both the saliency of health-related misinformation and the monetization of wellness content, we offer insight into the gendered dimension of misinformation spread. We then present our methods, drawing on a digital ethnography of three wellness influencers on Instagram. Ultimately, our analysis reveals how influencers take advantage of the platform’s sociotechnical infrastructure and attempt to profit from misinformation while normalizing antivaccine sentiment and conspiratorial rhetoric.

Fakes

'Truths and Tales: Understanding Online Fake News Networks in South Korea' by Benedict Sheehy, Sujin Choi, Md Irfanuzzaman Khan, Bruce Baer Arnold, Yoonmo Sang and Jae-Jin Lee in (2024) Journal of Asian and African Studies comments 

This study investigates the features of fake news networks and how they spread during the 2020 South Korean election. Using actor–network theory (ANT), we assessed the network’s central players and how they are connected. Results reveal the characteristics of the videoclips and channel networks responsible for the propagation of fake news. Analysis of the videoclip network reveals a high number of detected fake news videos and a high density of connections among users. Assessment of news videoclips on both actual and fake news networks reveals that the real news network is more concentrated. However, the scale of the network may play a role in these variations. Statistics for network centralization reveal that users are spread out over the network, pointing to its decentralized character. A closer look at the real and fake news networks inside videos and channels reveals similar trends. We find that the density of the real news videoclip network is higher than that of the fake news network, whereas the fake news channel networks are denser than their real news counterparts, which may indicate greater activity and interconnectedness in their transmission. We also found that fake news videoclips had more likes than real news videoclips, whereas real news videoclips had more dislikes than fake news videoclips. These findings strongly suggest that fake news videoclips are more accepted when people watch them on YouTube. In addition, we used semantic networks and automated content analysis to uncover common language patterns in fake news, which helps us better understand the structure and dynamics of the networks involved in the dissemination of fake news. The findings reported here provide important insights on how fake news spread via social networks during the South Korean election of 2020. The results of this study have important implications for the campaign against fake news and ensuring factual coverage.

Fake News

'Tackling online false information in the United Kingdom: The Online Safety Act 2023 and its disconnection from free speech law and theory' by Peter Coe in (2024) Journal of Media Law comments  

In the UK, there has been consistent recognition from a variety of actors, including the UK government, that the dissemination of false information can be harmful to individuals and the public sphere. It has also been acknowledged that this problem is being exacerbated by the role played in our lives by the likes of Google, Facebook, Instagram, and X, and because the systems that were in place for dealing with this type of content (and other illegal and/or harmful content), prior to the introduction of the Online Safety Act 2023 (OSA), were designed for the offline world, and were (and in some cases, still are) outdated and no longer fit for purpose. 

The UK’s online harms regime has intensified this debate. The regime began life in April 2019 as the Online Harms White Paper, morphing into multiple iterations of the Online Safety Bill (OSB), published in its original form in May 2021, and finally crystallising as the OSA, which was enacted on the 26th of October 2023. On the one hand, it is acknowledged that legislation placing statutory responsibilities on internet services to prevent the publication of false information (and other illegal and harmful content) may benefit society and public discourse. This is because, in theory at least, by helping to decrease the volume of false information we are exposed to, such laws should reduce the opportunities for the public sphere to become distorted. As citizens we should be able to assess, with greater confidence, the veracity of information available to us, and in turn, use this information, and the trust we have in it, to make positive contributions to public discourse. 

But, on the other hand, the OSA has been (and before it, the OSB was) met with significant resistance from a variety of actors because of the potential threats to free speech that it presents.  Indeed, since the publication of the White Paper, and the initial draft of the OSB, the regime has been shrouded in controversy. The OSB was subject to numerous amendments, and at one stage, it looked as though it would be scrapped altogether. Yet despite this, at the time of writing, the OSA has recently been enacted, albeit the overall shape of the regime remains unclear, because much of the legal detail will be contained in secondary legislation. Therefore, debates on the efficacy of the OSA will continue, and only time will tell what its ultimate impact on free speech will be. 

Notwithstanding this uncertainty, the purpose of this article is to interrogate the regime’s compatibility with free speech law and theory. In doing so, it begins with an explanation of what is meant by false information, and how the phenomenon has been exacerbated by the internet. This is followed by analysis of the pre-OSA system for dealing with this content, and an explanation of why it did not work, as aspects of it have a bearing upon the OSA regime. Next, the contours of the free speech framework are sketched, including relevant jurisprudence of the European Court of Human Rights (ECtHR), and the theories underpinning it that are particularly relevant to online false information. In this section I explain why these theories are flawed in this context, and therefore how these flaws could justify the creation of laws to tackle online false information. Yet, as I go on to suggest in my analysis of the OSA, which follows, this creates a paradoxical disconnect between theory and law, in that although the flaws in the theories may justify the creation of such laws – which manifests as the OSA – its creation arguably conflicts with the ECtHR’s jurisprudence, and the spirit of its theoretical foundations, and could inadvertently interfere with free speech. Finally, the article concludes with some potential solutions for meeting this challenge that do not erode one of the core fundamental human rights.

Disinformation

'The efficacy of Facebook’s vaccine misinformation policies and architecture during the COVID-19 pandemic' by David A David A Broniatowski, Joseph R Simons, Jiayan Gu, Amedia M Jamison and Lorien C Abroms in (2023) 9(37) Science Advances comments 

 Online misinformation promotes distrust in science, undermines public health, and may drive civil unrest. During the coronavirus disease 2019 pandemic, Facebook—the world’s largest social media company—began to remove vaccine misinformation as a matter of policy. We evaluated the efficacy of these policies using a comparative interrupted time-series design. We found that Facebook removed some antivaccine content, but we did not observe decreases in overall engagement with antivaccine content. Provaccine content was also removed, and antivaccine content became more misinformative, more politically polarized, and more likely to be seen in users’ newsfeeds. We explain these findings as a consequence of Facebook’s system architecture, which provides substantial flexibility to motivated users who wish to disseminate misinformation through multiple channels. Facebook’s architecture may therefore afford antivaccine content producers several means to circumvent the intent of misinformation removal policies. 

Online misinformation undermines trust in scientific evidence (1) and medical recommendations (2). It has been linked to harmful offline behaviors including stalled public health efforts (3), civil unrest (4), and mass violence (5). The coronavirus disease 2019 (COVID-19) pandemic has spurred widespread concern that misinformation spread on social media may have lowered vaccine uptake rates (6, 7). Therefore, policymakers and public officials have put substantial pressure on social media platforms to curtail misinformation spread (8, 9). 

Years of “soft” remedies—such as warning labels by Twitter (10), YouTube (11), and Facebook (12) and attempts by these platforms to downrank objectionable content in search—have demonstrated some success (13); however, misinformation continues to spread widely online, leading many to question the efficacy of these interventions (14). Some have suggested that combining these soft remedies with “hard” remedies (15)—removing content and objectionable accounts (16–18)—could largely curtail misinformation spread (19). However, evidence for the short-term efficacy of hard remedies is mixed (20–24), and the long-term efficacy of these strategies has not been systematically examined. Hard remedies have also spurred accusations of censorship and threats of legal action (25, 26). There is therefore a critical need to understand whether this combination of remedies is effective—i.e., whether it reduces users’ exposure to misinformation—and if not, why not. 

Any evaluation of the efficacy of these remedies must be grounded in a scientific understanding of why misinformation spreads online. Prior work indicates that misinformation may spread widely on social media if it is framed in a manner that is more compelling than true information (27, 28). Users appear to prefer sharing true information when cued to think about accuracy (13); however, the social media environment may interfere with peoples’ ability to distinguish truth from falsehood (29, 30). Other studies have suggested that social media platforms’ algorithms facilitate the creation of “echo chambers” (31, 32), which increase exposure to content from like-minded individuals. Accordingly, prior interventions have focused on altering the social media environment to either reduce users’ exposure to misinformation or inform them when content is false. However, recent evidence suggests that people use online algorithms to actively seek out and engage with misinformation (33). Therefore, on the basis of prior theory (34, 35), we examine how a social media platform’s “system architecture”—its designed structure that shapes how information flows through the platform (34)—enables antivaccine content producers and users to flexibly (36) establish new paths to interdicted content, facilitating resistance to content removal efforts. 

We analyzed Facebook because it is the world’s largest social media platform. In December 2020, when COVID-19 vaccines first became available, Facebook had 2.80 billion monthly active users (37). As of April 2023, this number had grown to 2.99 billion monthly active users (25). We therefore conducted an evaluation of Facebook’s attempts to remove antivaccine misinformation from its public content throughout the COVID-19 pandemic. 

Of primary interest were the following three research questions: (i) Were Facebook’s policies associated with a substantial decrease in public antivaccine content and engagement with remaining antivaccine content? (ii) Did misinformation decrease when these policies were implemented? (iii) How might Facebook’s system architecture have enabled or undermined these policies?

...Our findings suggest that Facebook’s policies may have reduced the number of posts in antivaccine venues but did not induce a sustained reduction in engagement with antivaccine content. Misinformation proportions both on and off the platform appear to have increased. Furthermore, it appears that antivaccine page administrators especially focused on promoting content that outpaced updates to Facebook’s moderation policies: The largest increases appear to have been associated with topics falsely attributing severe vaccine adverse events and deaths to COVID-19 vaccines.

Fake News

'Fake News and the Tax Law' by Kathleen Delaney Thomas and Erin Scharff in Washington and Lee Law Review (Forthcoming) comments 

The public misunderstands many aspects of the tax system. For example, people frequently misunderstand how marginal tax rates work, misperceive their own average tax rates, and believe they benefit from tax deductions for which they are ineligible. Such confusion is understandable given the complexity of our tax laws. Unfortunately, research suggests these misconceptions shape voter preferences about tax policy which, in turn, impact the policies themselves. 

That people are easily confused by taxes is nothing new. However, with the rise of social media platforms, the speed at which misinformation campaigns can move to shape public opinion is far faster now. The past five years have seen a dramatic shift in the landscape of false information, and scholars in a variety of disciplines, from law to psychology to journalism, have explored the increasing influence of fake news. 

Building on this burgeoning literature, this Article is the first to examine the incidence and impact of fake news on the tax law. We analyze a unique dataset of tax stories flagged as “false” or “untrue” by reputable, third-party news sources. We use this dataset to explore common themes in fake tax news, as well as the ways tax laws’ complexity contributes to spreading false information. We then offer recommendations for how tax administrators and policymakers can combat these misinformation efforts. Specifically, we argue that insights from the literature on fake news can and should inform how administrators disseminate true tax information to the public. Further, understanding what types of tax laws are easily misunderstood or subject to manipulation should inform substantive tax policy design.

DIGI code

'Australia’s DIGI Code: what can we learn from the EU experience?' by Edward Hurcombe and James Meese in (2022) 57(3) Australian Journal of Political Science 297-307 comments

Disinformation and misinformation on social media platforms are growing regulatory concerns for governments. In this paper, we examine Australia’s DIGI Code of Practice (CoP). The CoP is a voluntary set of commitments to address false content on platforms that is overseen by Australia’s digital media industry. It was established in response to a federal government directive. We evaluate the DIGI CoP by comparing it to the code it was partially based on: the EU’s Code of Practice on Disinformation (EU CPD). The EU CPD also attempted voluntary regulation, and yet the EU is now moving towards a strengthened CPD. While the DIGI CoP improves upon the original EU CPD, the original version had limited accountability measures and notable content exclusions. We conclude by discussing a recent review of the CoP, suggesting that Australia will also strengthen regulation in this area. Lastly, we address ongoing tensions around media freedom and accountability. 

 The authors argue 

In the last five years, disinformation and misinformation on social media platforms have become serious regulatory concerns. This is due to several recent high-profile developments, such as attempts by external state actors to influence domestic elections (Parlapiano and Lee 2018). The COVID-19 pandemic has also highlighted the public health risks of platform-based misinformation (Bruns, Harrington, and Hurcombe 2020a; Meese and Hurcombe 2020b), so much so that World Health Organisation declared the latter an ‘infodemic’ that rivalled the virus itself (United Nations 2020). In response, governments are increasingly playing a role in reducing false information and malicious behaviour on platforms (Meese and Hurcombe 2020b; Alemanno 2018). 

Australia’s recent ‘Code of Practice on Disinformation and Misinformation’ (DIGI 2021), developed by the domestic digital industry body DIGI in response to a government directive, is one outcome of this broader policy trend. The Code of Practice (DIGI CoP) is a voluntary set of commitments made by tech companies to reduce the propagation of dis- and misinformation, empower consumers to identify dis- and misinformation, increase the transparency of political advertising on platforms, and publicise measures combating dis- and misinformation. The DIGI CoP emerged out of the Australian government’s Digital Platforms Inquiry, which recommended that a disinformation code be introduced (ACCC 2019). The Australian government accepted this recommendation and directed major tech companies to develop a voluntary code of conduct. Following subsequent stakeholder engagement as well as input from the Australian Communications and Media Authority (ACMA), the code was expanded to also address concerns around misinformation and news quality (ACMA 2020; DIGI 2021). However, the effectiveness of the code remains unclear. In this article, we evaluate the DIGI CoP by comparing it to the code it was based on: the EU’s Code of Practice on Disinformation (EU CPD). The EU CPD, which major tech companies such as Facebook, Google, and Twitter signed onto in October 2018 (EC n.d.), also attempted voluntary regulation as a means of reducing misinformation and improving advertising transparency on platforms. And yet, the European Commission is now moving towards a ‘strengthened’ EU CPD, following concerns about weaknesses in the original code. Below, we examine the original EU CPD, discuss how the DIGI CoP improves upon the EU CPD, as well identify ongoing weaknesses in the DIGI Code. We then reflect on what recent developments in the EU means for the future of the DIGI CoP.

Disinfo

Alongside a commitment - so heartfelt - to introduce disinfo legislation in the 2nd half of this year as part of the 'Australian Code of Practice on Disinformation and Misinformation', the Communications Minister has released the June 2021 A report to government on the adequacy of digital platforms’ disinformation and news quality measures.

Unsurprisingly there is no engagement with disinformation/misinfo from members of the Government such as Craig Kelly. 

 The report states 

In December 2019, as part of its response to the Australian Consumer and Competition Commission’s Digital Platforms Inquiry, the Australian Government requested that digital platforms in Australia develop a voluntary code of practice to address online disinformation and news quality. The Australian Code of Practice on Disinformation and Misinformation1 (the code) was launched by industry association Digital Industry Group Inc (DIGI) on 22 February 2021. The code has since been adopted by 8 digital platforms – Google, Facebook, Microsoft, Twitter, TikTok, Redbubble, Apple and Adobe. 

The ACMA was tasked with overseeing the development of the code and reporting to the government on the adequacy of platform measures and the broader impacts of disinformation in Australia. Our report provides new consumer research on users’ experience of disinformation and misinformation on digital platforms and our assessment of the industry’s code. It also provides a range of findings and a number of recommendations for consideration by the government. 

The online propagation of disinformation and misinformation presents an increasing threat to Australians 

Over the previous 18 months, we have seen increasing concern within the community over the ‘infodemic’ of online disinformation and misinformation, particularly in relation to the real-world impacts of COVID-19. The propagation of these falsehoods and conspiracies undermines public health efforts, causes harm to individuals, businesses and democratic institutions, and in some cases, incites individuals to carry out acts of violence. 

To understand the scale and impacts of this issue in Australia, we undertook a mixed- methods study focused on COVID-19 misinformation. Key insights include:

> Most adult Australians (82%) report having experienced misinformation about COVID-19 over the past 18 months. Of these, 22% of Australians report experiencing ‘a lot’ or ‘a great deal’ of misinformation online. 

> Belief in COVID-19 falsehoods or unproven claims appears to be related to high exposure to online misinformation and a lack of trust in news outlets or authoritative sources. Younger Australians are most at risk from misinformation, however there is also evidence of susceptibility among other vulnerable groups in Australian society. 

> Australians are most likely to see misinformation on larger digital platforms, like Facebook and Twitter. However, smaller private messaging apps and alternative social media services are also increasingly used to spread misinformation or conspiracies due to their less restrictive content moderation policies. 

> Misinformation typically spreads via highly emotive and engaging posts within small online conspiracy groups. These narratives are then amplified by international influencers, local public figures, and by coverage in the media. There is also some evidence of inorganic engagement and amplification, suggesting the presence of disinformation campaigns targeting Australians. 

> Many Australians are aware of platform measures to remove or label offending content but remain sceptical of platform motives and moderation decisions. There is widespread belief that addressing misinformation requires all parties – individuals, platforms and governments – to take greater responsibility to improve the online information environment and reduce potential harms. 

Digital platforms have introduced a range of measures in response to the growth of disinformation and misinformation on their services In response largely to global concerns, digital platforms have introduced measures typically based on company-wide policies including:

> supporting third-party fact-checking organisations 

> proactively updating their policies to specifically address unique events, such as the COVID-19 pandemic and the 2020 US presidential election 

> investing in means to signal credible, relevant and authentic information 

> providing financial assistance and grants to news outlets, government and not-for- profit organisations to bolster the spread of credible information and news 

> increased detection, monitoring and enforcement action against groups and networks who use their services to spread disinformation and misinformation. 

Despite platforms’ mostly global approach to updating policies and implementing other actions, many measures have had an impact on Australian users.

> In 2020, Facebook removed more than 110,000 pieces of COVID-related misinformation generated by Australian accounts. 

> Between July and December 2020, Twitter removed 50 pieces of content authored by Australian accounts for contravening its COVID-19 misleading information policy. 

> In 2020, Google blocked 101 million advertisements globally for contravening its misrepresentation policies. 

> TikTok’s COVID-19 Information Hub was visited by over 292,000 Australians between November 2020 and March 2021. 

The above data shows that platforms are taking proactive steps to tackle disinformation and misinformation on their products and services. The introduction of an Australian industry code builds on these actions to codify actions, improve transparency, enhance consumer protections, and implement mechanisms to monitor their effectiveness. It also provides a framework to promote stakeholder collaboration and incentivise further actions by platforms to respond to a rapidly evolving online environment. 

Digital platforms have come together to develop a single outcomes-based code of practice with several important features 

It is extremely positive to see industry, steered by DIGI, come together to develop a single code of practice. A single code should promote a consistent approach by platforms and provide confidence in industry to manage the range of harms associated with disinformation and misinformation. 

DIGI ran a meaningful public consultation process in developing its draft code, which attracted a variety of submissions that clearly influenced subsequent changes. In particular, the scope of the code was expanded to cover misinformation as well as disinformation, a key piece of stakeholder feedback during the consultation process. The ACMA considers this is an improvement on the EU Code of Practice on Disinformation. The code adopts an outcomes-based regulatory approach that allows a range of platforms with different services and business models to sign up to the single code. Signatories are required to sign up to the objective of ‘providing safeguards against harms that may arise from disinformation and misinformation’ and may opt-in to other code objectives, such as disrupting advertising incentives and supporting strategic research. The code also provides signatories flexibility to implement measures to counter disinformation and misinformation in proportion to the risk of potential harm. Signatories must also report annually on the range of measures they will implement to achieve the objectives and outcomes. Importantly, the code also stresses the need to balance interventions with the need to protect users’ freedom of expression, privacy, and other rights. 

Our assessment identifies further improvements that should be made to the code’s scope and the clarity of commitments 

The ACMA has assessed the code to consider whether it has met the expectations set out by the government and has identified a range of improvements. 

In our view, the scope of the code is limited by its definitions. In particular, a threshold of both ‘serious’ and 'imminent’ harm must be reached before action is required under the code. The effect of this is that signatories could comply with the code without having to take any action on the type of information which can, over time, contribute to a range of chronic harms, such as reductions in community cohesion and a lessening of trust in public institutions. 

The code should also be strengthened through an opt-out rather than opt-in model. Signatories should only be permitted to opt out of outcomes where that outcome is not relevant to their service and be required to provide justification for the decision. 

The code is also limited in the types of services and products it covers. Private messaging is excluded, despite increasing concern about the propagation of disinformation and misinformation through these services, particularly when used to broadcast to large groups. Including messaging services within the code, with appropriate caveats to protect user privacy (including the content of private messages), would provide important consumer protections. 

We also consider improvements to the code should be made in relation to: > its application to news aggregation services > the treatment of professional news content and paid and sponsored content > the weight given to news quality as a key aspect of the government’s request to industry. 

The ACMA is also concerned that the code does not place an obligation on individual signatories to have robust internal complaints processes. This was an area of particular concern identified in the Digital Platforms Inquiry. 

The code includes commitments to establish administrative functions within 6 months of code commencement. As code administrator, DIGI will establish a compliance sub- committee, a detailed reporting guideline and a facility to address signatory non- compliance. However, these functions remain under development at the time of finalising this report. As a result, the ACMA has not been able to assess their effectiveness. DIGI and code signatories should consider changes to the code to address the matters identified by the ACMA in its review in February 2022. 

A clear and transparent measurement framework is critical to the effectiveness of a voluntary, outcomes-based regulatory model Signatories were required to nominate their code commitments and deliver an initial report under the code, providing information and data on the measures they have adopted under the code. 

Signatories’ reports provide a large range of information on the actions they have taken to address disinformation, misinformation and news quality, and their investments in collaborative initiatives. 

However, reports are heavily focused on platform outputs and lack systematic data or key performance indicators (KPIs) that would establish a baseline and enable the tracking of platform and industry performance against code outcomes over time. Reports also show inconsistencies in the interpretations of key code terms and in reporting formats. 

Platforms should move quickly to identify KPIs specific to their services and work together to establish industry-wide KPIs to demonstrate the effectiveness of the code as an industry-wide initiative. 

The ACMA recommends a number of actions by government to bolster industry self-regulatory arrangements 

The ACMA considers that it is still too early to draw concrete conclusions on the overall impact or effectiveness of the code. The code administration framework – including a detailed reporting guideline and mechanism to handle complaints – is not due for completion until late August 2021. The design and implementation of these elements will be key to the overall effectiveness of the code. 

Given these circumstances, continued monitoring is required and the ACMA recommends it provide government with another report on the code by the end of the 2022–23 financial year. This will provide sufficient time to assess the operation of the code administration framework and assess the impact of any changes arising from the February 2022 review of the code. As part of this report, the ACMA recommends it continues to undertake focused research on these issues. 

Initial signatory reports identify challenges in obtaining relevant data on platform actions in Australia. Providing the ACMA with formal information-gathering powers (including powers to make record-keeping rules) would incentivise greater platform transparency and improve access to Australia-specific data on the effectiveness of measures to address disinformation and misinformation. Information collected could also be used to identify systemic issues across the digital platform industry and inform future ACMA research. 

More formal regulatory options could be considered, particularly for platforms that choose not to participate in the code or reject the emerging consensus on the need to address disinformation and misinformation. The ACMA recommends that government provides the ACMA with reserve regulatory powers in relation to digital platforms – such as code registration powers and the ability to set standards. This would provide the government with the option to act quickly to address potential harms if platform responses are not adequate or timely. 

There are also opportunities for improved collaboration between government agencies, platforms, researchers and non-government organisations on issues relating to disinformation and misinformation. The ACMA recommends that the government should consider establishing a Misinformation and Disinformation Action Group to provide a mechanism to support future information sharing, cooperation and collaboration. 

The ACMA makes 5 recommendations to the government in its report. 

Recommendation 1: The government should encourage DIGI to consider the findings in this report when reviewing the code in February 2022. 

Recommendation 2: The ACMA will continue to oversee the operation of the code and should report to government on its effectiveness no later than the end of the 2022- 23 financial year. The ACMA should also continue to undertake relevant research to inform government on the state of disinformation and misinformation in Australia. 

Recommendation 3: To incentivise greater transparency, the ACMA should be provided with formal information-gathering powers (including powers to make record keeping rules) to oversee digital platforms, including the ability to request Australia- specific data on the effectiveness of measures to address disinformation and misinformation. 

Recommendation 4: The government should provide the ACMA with reserve powers to register industry codes, enforce industry code compliance, and make standards relating to the activities of digital platforms’ corporations. These powers would provide a mechanism for further intervention if code administration arrangements prove inadequate, or the voluntary industry code fails. 

Recommendation 5: In addition to existing monitoring capabilities, the government should consider establishing a Misinformation and Disinformation Action Group to support collaboration and information-sharing between digital platforms, government agencies, researchers and NGOs on issues relating to disinformation and misinformation.

Elections and Social Media

The First Interim Report of the Senate Select Committee on Foreign Interference Through Social Media, which quotes my submission, features the following recommendations: 

R 1  The committee recommends that the Australian Government clearly delegate lead accountability for cyber-enabled foreign interference to a single entity in government. 

R 2  The committee recommends that the Australian Government take a proactive approach to protecting groups that are common targets of foreign interference but are not classified as government institutions. 

R 3  The committee recommends that the Australian Government establish appropriate, transparent, and non-political institutional mechanisms for publicly communicating cyber-enabled foreign interference in our elections and review the processes and protocols for classified briefings for the Opposition during caretaker with respect to cyber-enabled foreign interference. 

R 4  The committee recommends that the Australian Communications and Media Authority's report into the functioning of the Australian Code of Practice on Disinformation and Misinformation be publicly released as a matter of priority. 

R 5  The committee recommends that the Australian Government publicly release the Electoral Integrity Assurance Taskforce's terms of reference. 

R 6  The committee recommends that the Australian Government establish clear requirements and pathways for social media platforms to report suspected foreign interference, including disinformation and coordinated inauthentic behaviour, and other offensive and harmful content, and formalise agency remits, powers and resourcing arrangements accordingly. 

R 7  The committee recommends that the Election Integrity Assurance Taskforce undertake an audit to assess capability relevant to detecting disinformation prior to the coming election and, further, that the Australian Government consider providing information about relevant capabilities and resourcing to this committee as appropriate to assist in our deliberations.

Disinfo

'How do you solve a problem like misinformation?' by Ryan Calo, Chris Coward, Emma S Spiro, Kate Starbird and Jevin D West in (2021) 7(50) ScienceAdvances comments

Understanding key distinctions between misinformation/disinformation, speech/action, and mistaken belief/conviction provides an opportunity to expand research and policy toward more constructive online communication. 

Trying to navigate misinformation about COVID, climate change, politics, and countless other topics can be overwhelming. This is true for the public, researchers, journalists, and policy-makers alike. As researchers dedicated to the study and resistance of misinformation, we often find ourselves in conversation with government officials and others trying to understand and address the phenomenon. To help illuminate the complexities of misinformation and to guide policy, we find three distinctions helpful: misinformation versus disinformation, speech versus action, and mistaken belief versus conviction. Failing to appreciate these distinctions can lead to unproductive dead ends; understanding them is the first step toward recognizing misinformation and hopefully addressing it. 

The first key distinction covers misinformation—erroneous or misleading information to which the public may be exposed, engage with, and share—and disinformation. Disinformation refers to a purposive strategy to induce false belief, channel behavior, or damage trust. Misinformation is usually discrete or standalone, as when a neighbor shares a false rumor or overhears a misleading exchange. Disinformation tends to take the form of a multifaceted campaign with a predetermined financial, political, or other objective. Disinformation campaigns blend orchestrated action and organic activity, relying on the participation of willing but unwitting online audiences. ... 

Fighting misinformation is about identifying and addressing misleading messages. It is conceivable that a machine learning system could help flag misinformation or that legislation could define it. However, fighting disinformation is another matter. It is an exercise in disentangling the motivations of the various actors, some innocent and sincere, others strategic. The warning signs for a disinformation campaign may, ironically, involve true information and reasonable opinion. This suggests a need for researchers to follow people and strategies, rather than individual content alone, and for legislatures to address the problem at the level of incentives. The best way to address a foreign disinformation campaign, however, may be diplomacy and economic sanctions rather than artificial intelligence or tort law reforms. 

A second distinction is the legal difference between speech and action. The U.S. Constitution protects free speech; however, it does not necessarily protect deceptive speech coupled with harmful action. This distinction potentially removes barriers to accountability for social media platforms that fail to address misinformation. Laws could require procedural safeguards and reporting about misinformation without censoring speech or treating Facebook or Google like a publisher. At a minimum, the assertive steps taken by technology companies to address coronavirus misinformation from warnings to outright deplatforming demonstrate potential methods to counter harmful content that has long plagued the internet. 

We can and should ask more of internet platforms to address the conditions that they helped create and profit from, but what does the law say? The U.S. Constitution prohibits the government from censoring speech, even if the speech is misleading. Federal law (Section 230 of the Communications Decency Act), meanwhile, immunizes platforms like Facebook and Google from liability for speech on their platforms that originate from sources outside the company. Yet, neither the Constitution nor federal law grants legal protections for harmful conduct just because the action involves speech. For example, in striking down the Stolen Valor Act in 2012, which penalized lying about receipt of the Congressional Medal of Honor, the Supreme Court afforded states the power to criminalize fraud based on such a lie, and indeed, Congress passed a new version of the Stolen Valor Act the next year, with a requirement that lying about the medal had to be for the purpose of material gain for it to be criminal. In other words, the government can regulate doing things, or failing to do them, even if those things involve speech. ... 

The final key distinction relates to the nature of belief itself, specifically, the difference between a mistaken belief and a conviction. We recognize that the distinction between belief and behavior is a subject of enduring interest in the social sciences. Indeed, one of our team’s primary research questions examines how exposure to misinformation translates into both belief and behavior. Yet, the distinction between beliefs held out of mistake and beliefs held out of conviction remains undertheorized in both the research literature and within policy circles.

Conspiracies

'Where the Earth is flat and 9/11 is an inside job: A comparative algorithm audit of conspiratorial information in web search results' by Aleksandra Urmana, Mykola Makhortykh, Roberto Ulloa and Juhi Kulshrestha comments 

Web search engines are important online information intermediaries that are frequently used and highly trusted by the public despite multiple evidence of their outputs being subjected to inaccuracies and biases. One form of such inaccuracy, which so far received little scholarly attention, is the presence of conspiratorial information, namely pages promoting conspiracy theories. We address this gap by conducting a comparative algorithm audit to examine the distribution of conspiratorial information in search results across five search engines: Google, Bing, DuckDuckGo, Yahoo and Yandex. Using a virtual agent-based infrastructure, we systematically collect search outputs for six conspiracy theory-related queries (“flat earth”, “new world order”, “qanon”, “9/11”, “illuminati”, “george soros”) across three locations (two in the US and one in the UK) and two observation periods (March and May 2021). We find that all search engines except Google consistently displayed conspiracy-promoting results and returned links to conspiracy-dedicated websites in their top results, although the share of such content varied across queries. Most conspiracy-promoting results came from social media and conspiracy-dedicated websites while conspiracy-debunking information was shared by scientific websites and, to a lesser extent, legacy media. The fact that these observations are consistent across different locations and time periods highlight the possibility of some search engines systematically prioritizing conspiracy-promoting content and, thus, amplifying their distribution in the online environments. 

Web search engines (SEs) are crucial information gatekeepers in contemporary high-choice information environments (Van Aelst et al., 2017) with internet users turning to them on a daily basis (Urman and Makhortykh, 2021). At the same time, as demonstrated by a mounting body of evidence, search results can be inaccurate or biased (Kay et al., 2015; Kulshrestha et al., 2017; Makhortykh et al., 2020; Noble, 2018; Otterbacher et al., 2017). Still, search outputs are highly trusted by people and can influence their opinions on matters ranging from commercial brands to elections (e.g., Fisher et al., 2015; Nichols, 2017). Thus, malperformance of SEs can cause societal problems by leading, for example, to the spread of misinformation or of racial stereotypes (Noble, 2018). 

While the explorations of bias in search results are increasingly common (see below), other forms of SE malperformance, in particular the one related to inaccurate search results, remain under-studied with a few notable exceptions (Bernstam et al., 2008; Bradshaw, 2019; Cooper and Feder, 2004). Unlike biased outputs, which tend to disproportionately amplify a particular point of view - e.g., by associating modern technology with Whiteness (Makhortykh et al., 2021a), - inaccurate outputs contain factually incorrect information (e.g., that the Earth is flat). Consequently, inaccurate outputs have higher potential for misinforming the users of SEs, which in some cases can pose a threat for their individual well-being as well as the society. It is particularly valid for outputs promoting conspiracy theories, which unlike other forms of incorrect or biased search outputs has so far received meager attention from the scholarly community. As shown by the ongoing COVID-19 crisis (European Commission, 2021), conspiracy theories diminish trust towards authorities and scientific community which can undermine societal cohesion and lead to radicalization, in particular at the time of crises. 

In this paper, we address the above-mentioned gap by investigating the presence of content promoting conspiracy theories in web search results through a systematic comparative algorithm impact audit. We rely on virtual agent-based infrastructure to systematically collect search outputs for six conspiracy theory-related queries on five most popular SEs across three locations and two waves (in March and in May 2021). Out of six utilized queries, three correspond to specific conspiracy theories (“flat earth”, “new world order”, “qanon”) - and are likely to be utilized by users interested in respective theories. Another three broadly refer to subjects about which many conspiracy theories circulate (“9/11”, “illuminati”, “george soros”) - and can be utilized by users broadly interested in related topics, without specific interest in conspiracy theories. We then conduct a qualitative analysis of all retrieved results to establish their stance on conspiracy theories (e.g., promoting/debunking) and their sources (e.g., social media or scientific websites), and compare our observations across locations and time periods. With this paper, we contribute, first, to the body of research on the spread of conspiracy theories through online platforms by analyzing their presence in web search results which were not studied in this context before; and second, to the literature on algorithm auditing and quality of information provided by web search engines. 

The rest of the paper is organized as follows: we first review the state of research on inaccurate and biased information in web search and on conspiracy theories online. Then, we build on this review to formulate concrete research questions and describe the methodology of our study in detail. Finally, we summarize our results and discuss their implications as well as the limitations of the current research.

One example of conspiracy claims is Attorney-General for the State of Victoria v Shaw [2012] VSC 334 dealing with a vexatious litigant. 

The Court states 

 [11] After setting out in considerable detail the various proceedings issued by Mr Shaw prior to 2007, Hansen J summarised the position as follows:

I make the following observations concerning the legal proceedings brought by the defendant. The defendant has brought a large number of criminal prosecutions in which he has made a range of the most serious allegations, including treason and perverting the course of justice, against numerous public officials. In all cases the charges have been struck out on the basis that the relevant Director of Public Prosecutions took over and withdrew the charges. I accept that by the nature of the charges, the circumstances in which they were laid, and the material provided by the defendant in support of them, it can be inferred that each charge was untenable and doomed to fail. In this sense, the proceedings were vexatious legal proceedings instituted without any reasonable ground. As to whether they were instituted habitually and persistently, I note that while not an invariable rule, there does emerge from the material a pattern whereby the defendant has brought criminal proceedings against those persons who have made decisions adverse to him. For example, upon taking over and discontinuing criminal prosecutions, both the Victorian and Commonwealth Directors of Public Prosecutions were themselves charged. After refusing the defendant’s grand jury application, the five members of the Court of Appeal were charged with criminal offences. And, after refusing to accept for filing an application for a grand jury, Master Cain was charged with criminal offences. These are just a few examples. And although there was some variation in the wording of the defendant’s allegations against those he charged, the substance of his allegations remained the same, namely their complicity in indictable offences relating to a Freemason conspiracy and/or constitutional improprieties. 

I turn now to the proceedings in which the defendant sought to challenge determinations of the Magistrates’ Court in relation to traffic offences. In essence the defendant alleged that the law under which he was charged was invalid because the Victorian constitution was invalid. He also raised allegations of a Freemason conspiracy. In each case the defendant’s proceeding was dismissed or struck out. The material demonstrates that these proceedings, instituted over at least six years, were vexatious and had no prospect of success. 

I now turn to the applications to summon a grand jury. This category is particularly significant, as it is apparent, both from an overview of the legal proceedings instituted by the defendant generally, and from what his counsel said during argument, that the defendant ultimately seeks to place before a grand jury his allegations that: (a) a Freemason conspiracy has corrupted the judiciary and the court process in Victoria; (b) there is currently an illegal conspiracy, already commenced in Western Australia, to fracture the Commonwealth of Australia and create a republic in its place; and (c) the Victorian Constitution is invalid and enacted without legal authority. ... 

Viewing the matter overall, I am of the opinion that the defendant has habitually and persistently instituted vexatious legal proceedings, without any reasonable ground. The allegations made by the defendant are of the most serious nature, yet completely lacking in substance. ...

[18] I do not propose to set out every matter addressed in oral submissions by Mr Shaw, but will identify some of those that made their mark with me.

• The removal of the oath of allegiance and the oath sworn by Australian lawyers is an act of treason on the part of at least the Attorney-General and possibly the Victorian Parliament. This was also said to be relevant to the affidavit sworn by Ms English on behalf of the Attorney-General. 

• In a similar way, the Courts and Tribunals Legislation (Further Amendment) Act 2000 of this State has shattered the whole structure of “our jurisprudence” by removing the oath of allegiance. The Attorney-General at the time the Act was introduced, Mr Rob Hulls, had acted in a treasonous way for his involvement in removing the oath of allegiance and subsequently declaring Mr Shaw a vexatious litigant “for exposing it”. 

• The Acts Amendment and Repeal (Courts and Legal Practice) Act 2003 (WA) had, by omitting references to the Crown, fractured the constitution. 

• Every politician, State or Commonwealth, from Western Australia is not qualified to sit since the passing of the Acts Amendment and Repeal (Courts and Legal Practice) Act 2003 (WA). 

• The Public Prosecutions Act 1994 (Vic), by s 51(3) which separated the office of the DPP from that of the Crown, was unconstitutional. 

• The lack of qualifications of Mr Andrew McGinty, the Western Australian Attorney-General in the year 2000, who subsequently allowed legislation to be introduced that “turned the whole jurisprudence of Australia upside down”. 

• That charges against 54 defendants (including the former Prime Minister Mr John Howard) are still pending before a grand jury. 

• That two UK judges should sit in Victoria to determine the constitutional points he has raised. 

• There are possibly two Victorian constitutions. 

• The Australia Act 1986 (Cth), at the instigation of the then Prime Minister, Mr Bob Hawke, and six premiers, is totally invalid and fraudulent, as is the sale of the Commonwealth Bank. 

• Mr Hawke, the Prime Minister responsible for introducing the Australia Act 1986 (Cth) continues to act illegally, as demonstrated by the front page of the Weekly Times of June this year, which showed a photograph of Mr Hawke. 

• The removal of the grand jury provisions by the Criminal Procedure Act 2010 (Vic) was, in effect, a scheme devised by the Victorian Attorney-General to avoid being prosecuted by Mr Shaw before a grand jury and amounted to perverting the course of justice.

[19] Mr Shaw also relied upon a number of matters contained in the notice given under s 78(b) of the Judiciary Act 1903 (Cth) to the Victorian Attorney-General. It included the following allegations.

• The existence of a criminal conspiracy to move the people of the Commonwealth into and under the United Nations, with or without the people’s knowledge and/or consent. 

• That the grand jury proceedings against the former Attorney-General, Mr Hulls, remained pending. 

• That in 2008, the then Attorney-General (Mr Hulls) perverted the course of justice when he introduced the Criminal Procedure Bill 2008 (Vic). 

• The decision of the Full Court in Re Shaw is “nugatory because the five judges entered into evidence and in doing so moved into the exclusive jurisdiction of a Grand Jury (23 electors) thereby voiding any ruling, order or judgment”. 

• That all State Governors are “Principal Offenders” and asserted that the Supreme Court, the County Court, the Magistrates’ Court and VCAT had their independent jurisdictions removed under Business Unit 19 of the Justice Department. 

• That the passing of the Western Australian Acts Amendment and Repeal (Courts and Legal Practice) Act 2003 had “broken unlawfully” the Act of Settlement 1700 (UK) and was an act of treason against both the Queen and the people. 

• That Judges of the High Court and the Governor-General were principal offenders and that the Premier, Deputy Premier or Attorney-General, Governor, Chief Justice and President of the Court of Appeal were each involved in a criminal conspiracy against the people. 

• That the principal issue in the Supreme Court in issuing a vexatious judgment related to and involved Freemasonry and the oaths and obligations of Freemasonry and the Masonic allegiance, and that the Supreme Court is a Masonic lodge. 

• That many lawyers, judges and magistrates are Catholic with no working knowledge either of scripture or the constitution.

[20] I do not think it is necessary to itemise any more of the allegations contained in this document, nor is it necessary to repeat the allegations contained in the written submissions which are in a similar vein.

The dastardly freemasons appear in other claims such as Haughton, Commonwealth Bank of Australia & Ors v Ridout & Ors [2004] WASC 136 and National Australia Bank v Walter [2004] VSC 36. In the latter the Court states 

The Walters raised a number of unorthodox arguments and challenges to jurisdiction at the commencement of the trial, on which I ruled at the outset. The matters in question included the alleged impact of Freemasonry, an alleged banking practice described as "fractional reserve banking", the invalidity of the Constitution of the State of Victoria, the Walters' entitlement to trial by jury under Magna Carta, and apprehended bias on my part, due to my disclosure of beneficial ownership of a parcel of NAB shares. 

I determined that none of the Walters' challenges to the Court's jurisdiction was of any substance. I also ruled that the issues of Freemasonry and fractional reserve banking were of no substance and irrelevant to any legitimate claim. Despite those rulings, the issues, which were not clearly defined, were persistently raised by the Walters in various altered guises throughout the course of the trial. 

The claims and challenges based on Freemasonry, fractional reserve banking, constitutional invalidity and Magna Carta which were raised by the Walters in these proceedings have previously been raised by litigants in person in the course of enforcement proceedings by banks. All have been the subject of some degree of previous judicial consideration and have been dismissed as wanting substance or as nonsense. Although those arguments occupied a considerable time at trial, the Walters also advanced a more conventional claim that the loans and securities were unenforceable on various related equitable grounds, including unconscionable conduct, duress, undue influence and estoppel. ... 

The Walters contended that the Court lacked jurisdiction to hear and determine the proceedings and was unlawfully constituted because certain judges and other Court officials are, or are suspected to be, Freemasons. They alleged that Freemasons administer and swear unlawful oaths, including oaths of allegiance to a foreign power, contrary to s.316 of the Crimes Act and s.321 of the Crimes Act. Further, the Walters contended that Freemasons are party to conspiracies to commit criminal acts and are otherwise implicated in criminal conduct. 

Ms Walter read to the Court some oaths allegedly administered to, and taken by, Freemasons. The Walters served a subpoena on an associate of a judge of the Court, requiring him to produce documentation which would reveal the identity of any judges, masters or other Court officials or employees who were Freemasons. 

The Walters contended that Freemasonry is a brotherhood of persons who habitually take unlawful oaths and who owe obedience to foreign powers. They alleged that in the course of their dealing with the NAB, Mr Fritz Walter (who is not a Freemason) failed to respond to a secret Masonic handshake made by an unidentified bank officer. The Walters claimed that in consequence, the NAB thereafter acted to the Walters' detriment and ultimately sold their property. No evidence of the alleged handshake incident was adduced at any stage. However, the Walters asserted that alleged Freemasonry within the Court precluded a fair trial of their claims. Ms Walter stated: "If the judge hearing the case were a Freemason, and the other party was a Masonic member as well and they had discussed the court case previously and made their decision while they were in the lodge" then "a litigant could not win." 

Master Evans, whom the Walters believed to be a Freemason (as he would neither confirm nor deny membership) had made an order in the principal proceeding for trial by judge alone. Master Evans' order was said to be of no effect, due to his alleged status as a Freemason. 

Although I stated that I was not, and had never been, a Freemason, the Walters contended that the status of the individual judge hearing the proceeding was irrelevant. They claimed that the Bench of the Supreme Court of Victoria was infested with Freemasons who were guilty of criminal acts, indictable offences and other unlawful conduct which contaminated the entire Court. Although Ms Walter preferred to characterise it as a "question", she essentially submitted that the Court lacked jurisdiction to determine the proceedings on the ground of Freemasonry. 

In reliance upon Re Shaw & Another, in which the Court of Appeal considered almost identical arguments about Freemasonry to those raised in the present matter, I ruled from the outset that Freemasonry had no bearing on any legitimate issue to be determined in the case. The reiteration of such allegations and the associated baseless attack on the Court's personnel and processes were, in my view, irresponsible and regrettable. SLAs where the High Court was unimpressed include Knight v Bell and Anor M43/2000 [2002] HCATrans 446 and Fyffe v The State of Victoria M123/1999 [2002] HCATrans 442.

Hacking

'The Simulation of Scandal: Hack-and-Leak Operations, the Gulf States, and U.S. Politics' by James Shires in (2020) 3(4) Texas National Security Review 10–29 comments 

Four hack-and-leak operations in U.S. politics between 2016 and 2019, publicly attributed to the United Arab Emirates, Qatar, and Saudi Arabia, should be seen as the “simulation of scandal" — deliberate attempts to direct moral judgment against their target. Although “hacking” tools enable easy access to secret information, they are a double-edged sword, as their discovery means the scandal becomes about the hack itself, not about the hacked information. There are wider consequences for cyber competition in situations of constraint where both sides are strategic partners, as in the case of the United States and its allies in the Persian Gulf. 

Hack-and-leak operations (HLO) are a new frontier in digital forms of foreign interference, epitomized by the success of Russian intelligence agencies in obtaining and disseminating documents from the Democratic National Committee (DNC) during the 2016 U.S. presidential election campaign.1 HLO and other information operations are widely seen as a severe threat to liberal democratic structures and U.S. policymakers have mobilized significant resources in response, including threat intelligence and cyber security protections, increased election and voting security, legislative pressure on social media companies, and even offensive cyber attacks. 

This “whole-of-nation” approach is largely based on the events of the 2016 U.S. election, and specifically Russian interference in the election process. However, it is hard to pinpoint the exact impact of the Russian disinformation operations. Controversial candidates, a combative and polarized media environment, and entrenched economic and social divisions were all key factors in the 2016 election result. Furthermore, foreign interest in the U.S. election was not limited to the Russian government; other state and nonstate actors also sought to influence candidate campaigns in their favor. The danger is that academic and policy understandings of HLO are over-reliant on a single case. This article therefore asks: How do other HLO cases alter our understanding of this new phenomenon, including motives, means, and consequences? 

HLO occur frequently worldwide, but their political contexts vary widely and have uncertain implications for U.S. politics.6 Consequently, this article expands our understanding of HLO through a detailed qualitative analysis of four operations that targeted political figures in the United States in the period following the DNC operation (October 2016 to January 2019), thus keeping the political and media environment constant as far as possible. These cases replicate many of the striking features of the DNC operation: access through phishing, the release of large collections of emails, publication in national media outlets, and even direct references to “DCLeaks,” the identity assumed by the Russian intelligence agencies to disseminate the DNC documents. These cases have been publicly attributed to governments in the Middle East, namely Qatar, Saudi Arabia, and the United Arab Emirates (UAE), and thus broaden conceptions of digital foreign interference to allies as well as adversaries. 

This article argues that HLO are the “simulation of scandal”: deliberate attempts to direct public moral judgment against their target. The success of HLO depends on the shifting power dynamic between the scandal-maker and the scandal-subject, referred to in Arabic as kāshif and makshūf, respectively. At the center of this dynamic are the digital technologies used to obtain and release secret information. These hacking tools are a double-edged sword, as their discovery often means the scandal becomes about the hack itself, not about the hacked information; in other words, the kāshif becomes the makshūf. These cases also highlight other overlooked aspects of HLO: the utility of “activist” cover, the involvement of new actors such as public relations (PR) agencies and law firms, and the leaker’s wary reliance on mistrustful relationships with traditional media. Finally, the article identifies wider consequences for cyber competition in situations of constraint where both sides are strategic partners. In such situations, HLO offer a powerful but indirect and unpredictable means of influence. 

The first section places HLO within the literature on cyber conflict and information operations. The second section draws on sociological accounts of mediatized and digitalized leaks to explore the simulation of scandal. The rest of the article concerns the four case studies: The third section provides an overview of each case; the fourth analyzes their coverage in prominent media outlets; and the fifth discusses reasons behind their differing effects. A conclusion places this discussion in a broader strategic context, highlights limitations, and suggests further work.

Influencers

The UK Advertising Standards Authority report Influencer Ad Disclosure on Social Media: A report into Influencers’ rate of compliance of ad disclosure on Instagram comments 

 With steadily rising complaints to the ASA about whether influencer ads are appropriately disclosed on social media, we proactively monitored a select group of influencers with whom we have previously been in contact about how to disclose content as advertising. We did this to establish their rate of compliance with rules which require ads to be obviously identifiable as such. 

The three week monitoring period focused on content on Instagram because complaints to the ASA about Influencer ad disclosure tended to be on this particular platform. The rules on ad disclosure apply to all platforms and media where influencers choose to engage in advertising. Our analysis of over 24,000 individual ‘Stories’, posts, IGTV and reels across 122 UK-based influencers revealed a disappointing overall rate of compliance with the rules on making it sufficiently clear when they were being paid to promote a product or service. 

We have therefore written to all of the Influencers we monitored and the main brands who had partnered with one or more of them. Brands are held equally responsible for failing to adequately disclose advertising content. 

We have also requested that the influencers provide an assurance of future compliance with the rules and will undertake monitoring to ensure that they comply – or face sanctions. 

What did we find? 

• Inconsistent disclosure across Stories - when a piece of ad content spans a number of consecutive Stories, unless it’s absolutely clear that this is part of the same posting, each Story must be disclosed as an ad. 

• Inconsistent disclosure across Stories, IGTV, Reels, posts – we noted instances where a post would be accurately disclosed as an ad but a corresponding Story, was not. 

• Visibility of ad labels – where Stories were labelled as ads, we noted labels were often in a small font, obscured by the platform architecture or otherwise difficult to spot; mainly due to being in a very similar colour to the background of the Story where it was placed. 

• Affiliate content is still an ad – we noted the use of #affiliate or #aff with no additional upfront disclosure; those labels are not likely to be enough on their own to disclose to users the advertising nature of the content. 

• Own-brand ads – Influencers should not rely on bios or past posts to make it clear to consumers that they are connected to a product. 

Both the CAP Code and the law require marketers to leave consumers in no doubt about when they are engaging with advertising, and this includes advertising on social media. This is not a new rule specific to social media, it is a rule that has been applied across all non- broadcast media via the CAP Code and the equivalent rules in the BCAP Code for broadcast media, for many years. The emergence of new platforms online has enabled new methods of incorporating marketing communications within and around editorial or other non-advertising content. However, our own Ad Labelling research and Ofcom research shows the difficulty that consumers have in distinguishing certain types of online ads from surrounding content. Clear and prominent disclosure is therefore essential; in short, it must be obvious to consumers before they read, ‘like’ or otherwise interact with content if it is a marketing communication. 

The ASA and CAP have invested significant resource and effort into helping the influencer marketing industry understand their responsibilities under the ad rules. On top of the numerous rulings the ASA has published on ad disclosure on social media, we have hosted an ‘Influencer Responsibility’ event and produced numerous pieces of advice online, including the ‘Influencers guide to making clear that ads are ads’ and a ‘cheat sheet’, specifically designed to help Love Islanders navigate the rules. 

In spite of the work already undertaken to help influencers stick to the rules, complaints to the ASA on this subject remain high: 2020 saw a 55% increase on 2019 in complaints received about influencers, from 1,979 to 3,144 individual complaints. 61% of those complaints in 2020 were about ad disclosure on Instagram. 

With influencer marketing continuing to grow as the popularity of social media apps increases due to consumers spending more time online, it is likely that the ASA will receive more complaints about this type of advertising. Our 2019 Annual Report demonstrated that complaints about influencer posts made up more than one quarter of all complaints about online media, with online media complaints making up just short of half of all the complaints made to the ASA in 2019. 

Influencer advertising is not new and whilst we’re heartened to see large elements of the industry maturing and disclosing ad content transparently, we continue to see far too many incidences of non-disclosure, which threaten to bring this marketing discipline into disrepute and breed distrust in consumers. Through easy to use CAP guidance for influencers on how to comply with the rules, there is simply no excuse for influencers not to make clear to consumers when content has been paid- for by a brand. 

The Code Rules 

The CAP Code applies in full to ads in all non-broadcast media, including digital platforms, such as Instagram, Facebook, YouTube, Snapchat, Twitter and TikTok. Many of these Code rules are underpinned by legislation, including the Consumer Protection from Unfair Trading Regulations 2008 (CPRs). If an influencer fails to sufficiently disclose that a post is in fact marketing, then not only are they breaking the CAP Code, they – and the brand they’re working with – may well be breaking the law. The relevant ad disclosure rules of the Code state: We consider that, in most cases, the use of #ad (or similar) is the clearest way of communicating the commercial nature of advertising content, though using a platform’s own disclosure tools, such as Instagram’s Paid Partnership tool, can also help to distinguish advertising from other content. 

Monitoring 

In September 2020, CAP undertook a three week monitoring exercise to review the Instagram accounts of 122 UK-based influencers to assess whether advertising content was being properly disclosed. The focus was on Instagram Stories because complaints made to the ASA were predominantly about this platform and this particular platform feature, but we did also consider posts, reels and IGTV. We also understood that ad spend on Instagram Stories has also been increasing significantly. 

Monitored content was considered very likely to be advertising when it included a “push” to consumers towards a brand. This might include explicit reference to a discount code or a “swipe up” message, for example. Or, where the content was otherwise clearly linked to a specific product or service via traditional advertising references to price, benefits of product/service (with no corresponding negatives), highlighting of a brand, etc. Years of experience regulating complaints about influencer content have helped us to identify when the content is very likely to amount to undisclosed advertising. 

Influencers were primarily chosen on the basis of having been previously contacted about non-disclosure of advertising by the ASA, either in response to a complaint or via our self- initiated, proactive engagements with them on the matter. 

As part of that work, we identified a significant number of Stories that were likely to be ads that had not been disclosed as such. We assessed over 24,000 Stories and compliance rates were far below what we would expect. We categorised nearly one in four Stories as marketing and considered that only 35% of them were clearly labelled and obviously identifiable as advertising. 

Ads were largely found to fall into three sectors: Beauty, Clothing and Leisure. This is perhaps unsurprising to those who interact with Instagram on a daily basis and see posts, in their feeds, promoting make-up, fast-fashion, luxury brands and various trips to hotels, spas, and theme parks. 

However, in terms of following the rules, no sector stood out as having an acceptable rate of compliance when it came to labelling ads. This suggests that neither influencers nor the brands are taking enough care to ensure consumers know when influencer marketing is occurring.