The Lowy Institute convened the Digital Threats to Democracy (DTD) Dialogue on 12 October 2022. This Dialogue was funded by the New South Wales Department of Premier in Cabinet and was a day-long, closed-door session that brought together a distinguished group of diverse subject matter experts, government officials and civil society stakeholders to examine intersecting digital challenges to democracy. The aim of the Dialogue was to foster connections across subject matter and policy areas in order to spark new ideas and more coordinated approaches to meet these challenges. To foster frank discussion, the session was conducted under Chatham House rules. Therefore the comments and recommendations made during the Dialogue and reflected in this report are not attributed. Additionally, the summary of the Dialogue and recommendations for future consideration should not be taken as endorsed or agreed upon by all Dialogue participants but rather are a reflection of the ideas and topics discussed.
The Dialogue was the cornerstone of a broader 12-month project that seeks to identify and examine the intersecting digital threats to democracy across four key areas: online disinformation, online hate and extremism, tech-enabled foreign interference and regulation of the digital sphere.
The Dialogue was structured according to these key themes and organised and hosted by Research Fellow and Project Director Lydia Khalil from the Transnational Challenges Program at the Lowy Institute. The Dialogue was divided into five concurrent panels that featured presentations by subject matter experts, followed by a moderated discussion between Dialogue participants. The Dialogue also included two keynote speeches delivered by international experts Nina Jankowicz, Vice President at the UK-based Centre for Information Resilience, and Dr Joan Donovan, Research Director of the Shorenstein Center on Media, Politics and Public Policy at Harvard University.
The following Summary Report consolidates and summarises the key points of the presentations, discussions and recommendations for consideration that arose from the DTD Dialogue.
On 12 October 2022, the Lowy Institute convened the Digital Threats to Democracy (DTD) Dialogue. The Dialogue brought together subject matter experts, government officials and civil society stakeholders to examine intersecting digital threats to democracy. The Dialogue was organised and hosted by Research Fellow and Project Director Lydia Khalil from the Transnational Challenges Program at the Lowy Institute. The aim of the Dialogue was to foster connections across subject matter and policy areas to spark new ideas and coordinated approaches to digital challenges to democracy.
The DTD Dialogue was structured around five panels that each featured presentations by subject matter experts, followed by a moderated discussion between Dialogue participants. The following are descriptions of the panel topics and issues considered.
Participants in the Dialogue examined and debated the challenges posed by and within the digital realm to the functioning of democratic procedures, levels of trust in democratic governance and the information environment that impacts the way citizens participate and interact in democratic societies. Two keynote speeches were delivered by international experts Nina Jankowicz, Vice President at the UK-based Centre for Information Resilience, and Dr Joan Donovan, Research Director
Panel presenters and Dialogue participants were asked to engage with how disinformation impacts citizens’ ability to access accurate information, which is essential for deliberation and decision-making in democracies. They also considered how disinformation is reducing trust in democratic governance, increasing polarisation, corrupting information ecosystems and even undermining consensus reality. A key question that Dialogue participants debated was what could be done to mitigate the spread of disinformation online or whether government should enact policies to counter disinformation online and its effects. The panel also assessed the criteria for what would make a successful countering disinformation program or policy.
A growing body of evidence demonstrates that the internet can be an important factor in facilitating radicalisation to violent extremism. At the same time, there is acknowledgement that such a broad conclusion requires more detailed analysis. The panel engaged with how the internet and other computer-mediated communications can have multiple and various roles in facilitating radicalisation and mobilisation to violent extremism. Discussion centred on whether content moderation was an effective or sufficient mechanism to counter the expression of violent extremism online and what else should be considered to counter online extremism and its real-world harms.
The digital environment has provided more opportunities for malign foreign influence and foreign interference. Through digitally enabled information warfare operations, election interference, deep fakes and various other means of undermining democratic political processes and institutions, foreign actors are violating national sovereignty via digital technologies. Participants discussed how democracies, in responding to this challenge, should react proportionately and according to democratic principles. The panel also addressed the ways in which digitally enabled disinformation, extremism and foreign interference are linked. They considered a wide range of comprehensive policy responses to address these interrelated digital challenges to democracy.
After many years of a laissez-faire approach to the tech sector, there are increasingly louder calls for tighter regulation — and government has responded. But despite the new regulations that are being enacted and considered, there are few that address the tech sector’s underlying business model of data acquisition and exploitation. Dialogue participants discussed the tensions between safety regulations and concerns about privacy and freedom of expression and how to best balance these competing priorities. Participants also considered regulations that would proffer greater transparency, particularly algorithmic transparency, from digital platforms and how gaining a greater understanding of how digital platforms function would help to address digital challenges to democracy.
In multicultural democracies and pluralistic societies, certain communities can be targeted as a means to undermine democratic institutions and social cohesion. At the same time, individual citizens and civil society groups have found ways to harness the digital environment to better engage in deliberation, dialogue and to address polarisation and other digital challenges. Dialogue participants examined ways in which particular communities have been impacted by online harms and how civil society and government can best mobilise to support solutions to these challenges.
In the process of robust discussion and dialogue, the DTD Dialogue generated a number of recommendations from participants. Below is a summary of those recommendations for consideration. These ideas for future consideration should not be taken as endorsed or agreed upon by all Dialogue participants.
- Disinformation or conspiratorial narratives spread online are often a hodgepodge of disjointed, even contradictory claims. These narratives do not need to make sense to their believers, rather individuals engage in disinformation and conspiracy theories to fulfill other psychosocial needs and to participate, coalesce and cohere around communities and social movements. Therefore, in order to address disinformation, actions beyond mere fact-checking and debunking campaigns must be used to counter damaging disinformation and conspiracy theories. Instead, governments and civil society actors must address the sociality of disinformation and conspiracy beliefs rather than their veracity.
- Government needs to communicate proactively, clearly and consistently with the public about its countering disinformation efforts. Democratic citizens are within their rights to question government efforts to influence or regulate discourse and behaviour. Therefore, governments need to clearly communicate why and when such actions are taken.
- It is also important to establish a threshold for when disinformation targeting government agencies or programs requires a response from government. Not all low-level disinformation will require a response — sometimes a response will only serve to amplify the disinformation. But when it does reach that identified threshold, governments should ensure that there is an agile and efficient response in place.
- Be prepared and be proactive. Government agencies and officials need to plan and have strategies ready for online malign foreign influence and disinformation campaigns targeting government and institutions. Government agencies and responsible civil society actors should project domain expertise so that the void is not filled by disinformation or other forms of mal-information.
- Prebunking has been shown to work more effectively than debunking mis- and disinformation narratives and campaigns. The way that social media platforms are currently designed gives advantage to first movers, so prebunking or information inoculation can be more effective in addressing the harms of disinformation and other forms of mal-information.
- There needs to be a greater focus on building citizen resilience to disinformation and other online harms rather than relying primarily on content moderation and counter-disinformation campaigns.
- It is important to go where the people are — fact sheets on government websites are insufficient as often people may not go to official government agency websites as the first port of call to obtain information. Government communications campaigns need to incorporate concurrent opportunities to engage on social media and legacy media, and via both online and offline local community organisations and hubs.
- Creating disinformation registers can highlight and help debunk disinformation campaigns and narratives. Disinformation registers can also serve as important resources for researchers and analysts.
- Public interest journalism is an effective antidote to disinformation and other forms of mal-information. Providing more awareness and training for journalists can be an effective means of countering the spread and harmful effects of disinformation. It is also important to provide awareness for journalists on how legacy and mainstream media can inadvertently spread and amplify disinformation and other harmful content.
- Unfortunately, the more the issue of disinformation is raised, the more distrust is potentially engendered among the public around official sources of information and mainstream news. One suggested work-around is to encourage the active consumption of information (i.e. asking who is writing it and who is funding it.)
- Generalised civics education can play an important role in addressing these intersecting challenges to democracy. Educating the public on the functioning of parliamentary democracy, levels of government, the functioning of bureaucracies, elections and representation may help buffer disinformation around political power and authorities.
- Government should back and defend public-facing civil servants and public institutions, proactively safeguarding their reputation and integrity instead of reactively responding to crises or attacks.
- Government agencies should build their reputation for the long term by building a track record of engagement and trust with the public. This will lend greater credibility to government communications when officials or agencies need to respond to a major event or crisis or to counter disinformation. They must be continuously engaging in the information space rather than reacting when issues arise.
- It is possible to reduce support for undemocratic candidates and reduce polarisation using short and scalable online interventions, but there is no one-size-fits-all approach and different causes require different interventions. The most successful online interventions have involved reducing tolerance for undemocratic practices and strengthening support for democratic principles. Other successful interventions have focused on reducing or correcting anti-democratic misperceptions of political opponents. Further successful online interventions included those aimed at decreasing dislike for political opponents and addressing bias evaluation of politicised facts through the cultivation of joint or uniform identity among the citizenry.
- Harnessing and encouraging the power of civil society is a key approach that should be utilised more often by democratic governments and societies. Civil society organisations (CSOs) that address digital challenges to democracy are able to keep an appropriate distance from government, which helps their credibility and creates organic synergies. Working with CSOs can also assist government agencies in outreach efforts. However, these efforts are resource-intensive and often underfunded. Government can play a role by funding or working in coordination with these CSO efforts.
- Governments are well versed in citizen consultation and engagement. However there should be consideration for governments to actively pursue further opportunities for shared decision making. This can include considering deliberative democracy and participatory democracy models as a method of engendering trust and engagement with democracy.
- The country agnostic approach to public discussion and government strategies hampers a risk management-based approach to addressing FIO by non-government actors. Consideration should be given to adjusting this country agnostic approach in favour of identifying the countries from which FIO are coming from in order to more efficiently and appropriately allocate resources to manage the associated risks.
- Countering foreign interference (CFI) strategies must also manage social cohesion risks as more forward leaning CFI approaches could result in perverse outcomes for impacted communities.
- Investment should be made in community-level understanding to help address the challenge of FIO. Public engagement, public education and empowering decentralised responses are important ways to counter FIO. A risk mitigation rather than risk elimination approach that incorporates these greater public engagements would harness the strengths of democratic societies and structures. • Government should consider a national public facing counter-foreign interference strategy, just as government has done with its national counterterrorism strategy. There are well-established cross-jurisdictional structures to deal with other national security threats, such as terrorism, and they could be similarly applied to addressing malign FIO and foreign interference.
- The following principles could effectively guide Big Tech regulation: (1) expand regulation to include mitigation of risks from platform systems and processes; (2) expand regulation to include addressing risks and harms to community and society in addition to risks and harms to individuals; (3) ensure platform accountability and transparency rather than the current setting, which places the burden of responsibility on individual actors; (4) work towards comprehensive regulation that addresses gaps in the regulatory framework; (5) move away from self-regulation, self-reporting, voluntary transparency reporting and voluntary codes of conduct and instead move towards co-regulation and/or enforced/mandated regulation; and (6) resource and join up government regulators.
- Government could consider potential pathways for developing and funding more public digital infrastructure. Much in the same way there is publicly funded broadcasters, publicly funded public service digital spaces could potentially help mitigate the digital threats to democracy examined in this dialogue.
- Independent civil society and/or academic research audits of social media platforms can serve an important function to address platform risks and digital threats to democracy.
- Mainstream social media platforms maintain that they are not publishers and are therefore not liable for content on their platforms, claiming it is the individual users who post content that are individually liable. This removes the onus of responsibility from digital platforms. One potential approach would be to introduce a duty of care provision for digital platforms to reduce harms and threats to democracy.
- Extremism will always be a contested concept, whereas dehumanisation is a more easily defined and understood one. Addressing harmful online content and behaviour through this dehumanisation lens would be one way to disrupt the challenges and limitations of programs and policies that aim to combat extremism. Using the dehumanisation rather than extremism paradigm could also allow for more pre-emptive rather than reactive responses and address these harms in a way that increases and maintains social cohesion.
- Working across international jurisdictions and likeminded democracies is critical as most digital platforms in use today are multinational private companies headquartered outside Australia. Domestic efforts need to be supplemented and linked to international efforts among likeminded democracies.
The Digital Threats to Democracy Project is supported by the New South Wales Department of Premier and Cabinet. The information, advice and/or views expressed in this project are those of the project author/s and participant/s and do not necessarily reflect the views of the Lowy Institute or the NSW government.
In August 2024, in an episode of Lowy Institute Conversations, Lydia Khalil spoke with Australian eSafety Commissioner Julie Inman Grant, who leads the world’s first government regulatory agency committed to keeping its citizens safer online. Listen to the podcast here.
In February 2024, the Lowy Institute published a Policy Brief by Lydia Khalil, entitled "Overcoming digital threats to democracy", which explores how using deliberative democracy can enhance trust and legitimacy in digital spaces. Read the paper here.
