Skip to content

Canadian Democracy @ Work: Framing Paper

July 2024

Purpose of this Framing Paper 

The Dais has launched a major new initiative: Canadian Democracy @ Work. It is designed to combat the effects of disinformation and to increase democratic engagement among Canadians through accessible workplace training in civic, media, and AI literacy. The structure and dissemination of the training will be developed in partnership with leading Canadian employers. 

This framing paper is intended to guide the discussions that will inform the development of the training materials and program. The paper proceeds in four sections. 

Sections one and two serve as background, articulating the thinking and research that underpin the rationale for the project. The first section lays out the major threats to Canadian democracy, with emphasis on the challenges presented by disinformation and AI technologies. The second section explains why Canadian employers, specifically, should be actively engaged in addressing the challenges facing democracy. 

Sections three and four will be used to structure the discussions to take place with participating employers. The third section offers a detailed view of the project and the immediate aims of the training material. It also includes a menu of possible topics for the training sessions, and of possible delivery formats. The fourth section outlines the agenda for the initial roundtable discussions, and lists guiding questions and considerations for participants to reflect on before the sessions. 

1. Disinformation, Disengagement, and Democratic Decline

Manipulation and deception have always played a role in the information ecosystem and public life. Recently, however, new technologies have made falsehood particularly abundant and influential—to the point that some have begun to characterize our present as the “disinformation age.”1 Indeed, the effects of disinformation—the deliberate spread of false information—are already pronounced in the Canadian context. A representative survey conducted by the Dais in 2024 found that one in ten Canadians have a high degree of belief in disinformation, failing to correctly identify more than 25% of a range of falsehoods such as false narratives about vaccines, climate change, immigration, the Holocaust and the Russian invasion of Ukraine.2 The problem is likely to only get worse. 

The AI Challenge

The launch of social media platforms two decades ago established the possibility of unprecedented connectivity, but has increasingly enabled the rapid spread of disinformation on a global scale.3 The integration of bots on online platforms allowed for the inauthentic amplification of malign content, further increasing the reach of conspiracy and propaganda content.4 While the spread and influence of online disinformation has proven extremely difficult to counteract, new developments in artificial intelligence (AI) are poised to make this task much more difficult. 

AI technologies have already created new and highly compelling forms of disinformation. Most notably, AI technologies have given rise to sophisticated and exceptionally life-like deepfakes. Deepfakes are synthetically altered videos in which the object or person depicted has been digitally modified to appear or sound like something or someone else.5 The introduction of deepfakes into the media ecosystem has made the authenticity of audio and video content—traditionally considered to be irrefutable proof—dubious. This has the twofold effect of easily convincing people of information that is untrue, for example when a German energy firm transferred nearly £200 000 after receiving a phone call from a fraudster who used AI technology to mimic the CEO’s voice,6 and of giving those in positions of power unconditional plausible deniability, allowing them to dismiss any evidence as doctored, as when President Trump falsely claimed that the infamous tape of him describing sexual assault was faked.7

Until recently, high-end deepfakes had intensive money, time, and computing demands.8 This is quickly changing with the dawn of generative AI, as creating fake images, audio, and videos is becoming cheaper and easier—meaning deepfake content could fundamentally undermine journalism and trustworthy sources of information.9 There are also so-called “cheap fakes,” which consist of simpler AV manipulation that can be rendered through Photoshop, clever editing, and other cheap software tools, are already easily accessible and have already destabilized traditional rules and norms around evidence, truth, and institutional authority.10

While the threat to the integrity of information posed by deepfake technology is relatively obvious, AI-generated text also presents new issues. Drawing on natural language processing, AI can generate authentic-sounding text, creating the possibility for original, text-based propaganda posing as journalism or science to be produced at scale.11

Disinformation and Democracy 

Disinformation drives political cynicism and disengagement, which presents major challenges for democratic institutions. The causal link between disinformation and political cynicism has two dimensions. First, disinformation tends, overwhelmingly, to be negative.12 Especially ahead of and during election periods, disinformation casts politicians as purely self-interested and the entire democratic system as ineffective or corrupt.13 Empirical studies show that exposure to such content promotes political cynicism.14 While the specific content of disinformation erodes trust in politics, the existence of disinformation itself, and the uncertainty it creates about the integrity of information and about the function of political institutions, also undermines confidence in the democratic process.15

Political cynicism decreases feelings of political efficacy and trust in democratic institutions.16 These are prerequisites for democratic participation of any sort. For instance, empirical analysis confirms high correlation between political trust and voter turnout.17 For people to participate in politics, they need to be able to trust that they can understand their political context, and that they can meaningfully affect it. Disinformation, by contributing to political cynicism, disincentivizes democratic participation, ultimately hindering the core functions of democratic institutions—namely, representing and serving citizens’ interests. 

2. Why Businesses Need Democracy

In response to the overlapping crises facing democracies around the world, calls to protect it have grown. Multilateral organizations, non-profits, schools and governments have—to various degrees—heeded the call, launching initiatives ranging from education programs and curriculum changes, to election integrity efforts and changes to legislation. 

Notably, to date, Canadian workplaces have not played a significant role in pro-democracy efforts. This is a missed opportunity. Working-age Canadians spend the largest portion of their waking hours at work.18 They also have a relatively high degree of trust in their employers. A 2021 survey of Canadians found employers to be highly trustworthy, with survey respondents citing employer communications as the most believable source of information, ahead of the national government, media reports, and social media. A survey from 2023 found employers to be a mainstay of trust, even as trust in other core institutions fell, with 75% of respondents reporting that they felt that their employer was trustworthy.20

Canadian employers have the opportunity to take on a substantive role in addressing this challenge—including the rising tide of disinformation and growing democratic disengagement. And it is in their best interest to do so. 

Business Costs of Disinformation 

Canadians’ difficulties in differentiating true content from false presents serious problems for democratic institutions. It is also, increasingly, cause for concern for businesses. Companies are often the targets of disinformation. One recent analysis of disinformation, for instance, found that companies were amongst the groups most frequently victimized by false, reputation-harming content, preceded only by political parties.21 Most of the disinformation targeting companies involved fake give-aways and job offers, which undermine the companies’ credibility and reputation.22 There were also cases of disinformation that targeted company spokespeople or representatives.23

The influence and reach of disinformation that targets companies is considerable. The fact checking website Snopes keeps a list of the most-searched rumours. This list has included false claims that United Airlines was filing for bankruptcy, that Starbucks was distributing free frappuccinos to undocumented workers, and that the CEO of PepsiCo had told supporters of President Trump that they could “take their business elsewhere.”24 Such claims have personal costs, especially in cases when an individual spokesperson is the target. They also have financial costs. An analysis of the effects of disinformation on the stock returns of targeted firms established that negative false news items have negative and significant short-term effects on returns.25

Disinformation, broadly, then, presents a threat to business success. Credence in disinformation among employees also has significant financial consequences for businesses. In May 2024, the British engineering firm, Arup, confirmed that it was victim to a deepfake scam, during which an employee was duped into sending £20 million to a criminal organization by an AI-generated video call.26 Also in May 2024, the Guardian reported that the CEO of WPP, the world’s biggest advertising group, was targeted by deepfake fraud which used an AI-generated voice clone.27 Indeed, scholars and business leaders alike anticipate the rates of fraud will increase in the coming years, especially as AI becomes more sophisticated.28

When employees fall victim to disinformation, the financial repercussions are significant. In a survey of Canadian business leaders whose organizations have faced fraud, 53% of respondents reported that their company had lost between 1% and 5% of their profits on account in the past 12 months.29 Only 4% of respondents reported that fraud had not had an impact on their returns.30 For more on how disinformation is targeting Canadian companies, see the Dais’ recent white paper on this topic with the Rogers Cybersecure Catalyst and RBC.

Business Costs of Democratic Decline 

Given the specific costs of disinformation on businesses, it is in their best interests to help mitigate its effects, even without taking into account its democracy-eroding effects. Businesses, however, should also be looking to shore up democratic institutions and processes. Reports from the World Bank and from Freedom House draw correlations between stable, transparent governments and operating environments that are conducive to business growth.31 Countries that performed well in assessments of respect for human rights, the rule of law, and democratic norms, had better economic and business outcomes. Conversely, countries with non-democratic governments were also more likely to impose red tape and trade barriers and to fail to enforce contracts, creating an inhospitable operating environment for businesses.32  

Extensive research points to the mutually-reinforcing relationship between democracy and the market economy. A 2019 paper titled “Democracy Does Cause Growth,” concluded that there is an economically and statistically significant positive relationship between a well-functioning democracy and long-run GDP per capita.33 Research by the V-Dem Institute found that democracies have lower variation in economic growth and are less likely to experience economic crises.34 Martin Wolf, the chief economics commentator for the Financial Times, attributes this to the built-in safety mechanisms that characterize democratic regimes—the most foundational of which is free and fair elections.35

A functioning democracy, then, ought to be understood as the base condition for prosperity, and for business success. As the prerequisites for a functioning Canadian democracy—information literacy, civic knowledge, and civic engagement—falter, Canadian businesses have both the reasons and the means to bolster them. 

3. Canadian Democracy @ Work

This project engages leading Canadian workplaces in the fight to protect Canadian democracy. This project aims to pilot providing neutral and non-partisan information literacy and civics education through Canadian workplaces in the public and private sector to equip working-age Canadians with the skills and knowledge they need to be effective and informed participants in the democratic process. The output of this project will be an accessible and customizable platform through which digital, AI, and civic literacy can be taught at scale to Canadian adults. 

There is evidence that media literacy education is effective in producing more critical, engaged, and ethical creators and consumers of media. Pre- and post-tests of students who participated in media literacy programs have generally shown that the programs succeeded in improving students’ capacity to accurately and effectively evaluate the media they consume, and in connecting students’ critical thinking with behavioral change—including increased civic participation.36 There is, likewise, evidence that civics education is positively associated with democratic participation.37 A systematic analysis of political participation and education across 23 countries concluded that civics courses play a significant role in inculcating a sense of civic duty, and in driving civic participation.38

Despite the promising results of both civics and media literacy programming, their reach remains limited. Learning about democratic and civic engagement in Canada occurs almost exclusively in the K-12 setting. Digital media literacy is just beginning to be taught in schools, and there are currently a patchwork of approaches across the country.39 For working-age Canadians, there is little meaningful and formalized education that enhances and refreshes their civic knowledge, or which equips them to navigate the current, algorithm-driven information landscape.

The design of this project takes inspiration from a similar initiative in Germany. Across Germany, several hundred companies have facilitated their employees’ participation in workshops on civics and democratic principles offered by groups such as the Business Council for Democracy and Weltoffenes.40 The workshops—which cover content ranging from recognizing disinformation, conspiracy theories, and hate speech to the importance of voting—are intended to improve employees’ well-being and job performance, while simultaneously strengthening German democracy.41 Several US-based organizations also administer similar workshops for working-age Americans.42 This project will be the first of its kind in Canada. 

The desired impacts of the project are as follows: 

  1. Employees will develop the ability to discern disinformation, with improved resistance to manipulation.
  2. Employees will have a better understanding of their own media consumption and a stronger sense of reliable information sources.
  3. Employees will feel more empowered to productively engage in conversations about difficult topics.
  4. Employees will renew their civics education and experience increased trust in democratic processes, resulting in benefits such as active participation in democratic processes and in civic life. 

The theory of change that underpins this project, and its big-picture objectives are well-established. The specific content and delivery approach of the trainings remain to be determined, based on employer insights and needs. Below are listed potential topics for the training and possible formats for these sessions:

Potential Training Content

  1. Identifying and combating disinformation 
    Participants would develop an understanding of common sources of disinformation and prevalent false narratives, learning how to analyze information and its sources. They would also assess their own media consumption habits and formulate strategies to combat disinformation in their personal and professional lives.
  2. Algorithms and AI 
    This training module would explore the role of AI and algorithms in spreading disinformation, highlighting how these technologies can manipulate information flow and public perception. Participants would learn to identify the mechanisms through which algorithms can amplify false information and the techniques used to create and spread AI-generated disinformation. The module would cover strategies for mitigating the impact of these technologies, including tips for identifying AI-generated content.
  1. Media literacy 
    This training module would equip participants with the skills to critically evaluate information from various media and news sources. Learners would explore how media messages are constructed and the impact they have on public opinion and behavior. The course would also teach participants strategies for discerning bias, verifying facts, and evaluating information in different media formats.
  1. Privacy, safety, and security on the internet
    This training module would provide participants with skills and knowledge they need to navigate online environments securely. Attendees would learn about privacy settings, safe browsing practices, and how best to protect their personal/work information from cyber threats. The course would also give participants a stronger understanding of their rights and responsibilities online.
  1. Civics basics 
    This module would furnish participants with an understanding of democratic institutions and processes in Canada. Topics covered would include the mechanics and history of Canada’s parliamentary system, the Constitution and the Charter of Rights and Freedoms, and Canadian federalism and the respective roles of the levels of government. 
  1. Elections 
    This module would give learners a comprehensive overview of voting systems, election laws, the history of the suffrage, and the significance of electoral participation. 
  1. Civic participation 
    This module would improve participants’ understanding of the relationship between democratic government and civic participation. The module would include a historical perspective on how citizen participation has shaped Canadian society; examples of civic participation and suggestions for ways to get involved; and an exploration of the skills, habits, and attitudes necessary for productive engagement in the political process. 
  1. Civil discourse 
    This module would equip participants to have healthy, constructive, and useful conversations about controversial or political topics. The module would cover principles of respectful civil discourse, as well as practical strategies for engaging in productive civic disagreement.

Potential Training Formats

  1. E-learning modules 
    Self-paced online training that would include interactive content like quizzes and videos to engage learners and track completion.
  2. Online webinars
    Pre-recorded sessions that could be accessed at any time and from anywhere. 
  3. In-person workshops
    Trainer-facilitated sessions would allow for real-time interaction, hands-on activities, and relationship-building. 
  1. Blended learning
    A combination of e-learning modules and in-person, facilitated discussions, allowing for both depth of engagement and flexibility 
  1. Micro-learning 
    A greater number of very short (5-10 minute) e-learning modules, focussed on a single skill or concept

4. Roundtable Agenda

The intention of these conversations is to gather insights to inform the design of the training modules to ensure they are as relevant and effective as possible. The purpose is to get insight, specifically, into the optimal content and delivery of the training modules. 

This roundtable has four portions: 

  1. A 15-minute introduction to the project and its aims
  2. A 40-minute discussion of what topics should be covered in the training modules 
  3. A 40-minute discussion of the ideal length and format of the training modules 
  4. A 15-minute discussion of next steps

Prior to the roundtables, participants are asked to reflect on the following questions:

  • What follow-up actions would you recommend to ensure the training has a lasting impact?
  • What training formats work best within your organization? 
  • How long should each training session be to ensure it is both comprehensive and engaging?
  • What topics do you believe are critical to include in the training modules?
  • What democracy-relevant skills are most important for your workplace? 
  • What resources does your organization have that could facilitate pro-democracy training?
  • What metrics does your organization typically use to measure the effectiveness of trainings? 
  • How can we tailor this content to be directly applicable to the specific needs of your organization?

1

Peter Pomerantsev, “The Disinformation Age: A Revolution in Propaganda,” The Guardian, July 27, 2019, https://www.theguardian.com/books/2019/jul/27/the-disinformation-age-a-revolution-in-propaganda.

2

Angus Lockhart, Mahtab Laghaei, and Sam Andrey, rep., Survey of Online Harms in Canada (The Dais, forthcoming).

3

Spencer Feingold, “Four Key Ways Disinformation Is Spread Online,” World Economic Forum, August 9, 2022, https://www.weforum.org/agenda/2022/08/four-ways-disinformation-campaigns-are-propagated-online/.

4

Ibid.

5

Todd C. Helmus, rep., Artificial Intelligence, Deepfakes, and Disinformation: A Primer (RAND Corporation, July 6, 2022), https://www.rand.org/pubs/perspectives/PEA1043-1.html, 2.

6

Catherine Stupp, “Fraudsters Used AI to Mimic CEO’s Voice in Unusual Cybercrime Case,” The Wall Street Journal , August 30, 2019, https://www.wsj.com/articles/fraudsters-use-ai-to-mimic-ceos-voice-in-unusual-cybercrime-case-11567157402.

7

Lauren Vadnjal, “How Deep Fakes Could Ruin Your Business,” Australian Financial Review, August 7, 2020, https://www.afr.com/technology/how-deep-fakes-could-ruin-your-business-20200804-p55id3.

8

Todd C. Helmus, rep., Artificial Intelligence, Deepfakes, and Disinformation: A Primer (RAND Corporation, July 6, 2022), https://www.rand.org/pubs/perspectives/PEA1043-1.html, 3.

9

Christoffer Waldemarsson , issue brief, Disinformation, Deepfakes & Democracy (Alliance of Democracies, April 27, 2020), https://www.allianceofdemocracies.org/wp-content/uploads/2020/04/Disinformation-Deepfakes-Democracy-Waldemarsson-2020.pdf, 11.

10

Britt Paris and Joan Donovan, “Deepfakes and Cheap Fakes,” Data & Society, September 18, 2019, https://datasociety.net/library/deepfakes-and-cheap-fakes/.

11

Tom Simonite, “To See the Future of Disinformation, You Build Robo-Trolls,” Wired, November 19, 2019, https://www.wired.com/story/to-see-the-future-of-disinformation-you-build-robo-trolls/.

12

S Mo Jones-Jang, Dam Hee Kim, and Kate Kenski, “Perceptions of Mis- or Disinformation Exposure Predict Political Cynicism: Evidence from a Two-Wave Survey during the 2018 US Midterm Elections,” New Media & Society 23, no. 10 (July 20, 2020): 3105–25, https://doi.org/10.1177/1461444820943878.

13

Ibid.

14

Joseph N. Cappella and Kathleen Hall Jamieson, Spiral of Cynicism: The Press and the Public Good (New York: Oxford University Press, 1997);
Matthijs Elenbaas and Claes H. de Vreese, “The Effects of Strategic News on Political Cynicism and Vote Choice among Young Voters,” Journal of Communication 58, no. 3 (September 2008): 550–67, https://doi.org/10.1111/j.1460-2466.2008.00399.x.

15

S Mo Jones-Jang, Dam Hee Kim, and Kate Kenski, “Perceptions of Mis- or Disinformation Exposure Predict Political Cynicism: Evidence from a Two-Wave Survey during the 2018 US Midterm Elections,” New Media & Society 23, no. 10 (July 20, 2020): 3105–25, https://doi.org/10.1177/1461444820943878.

16

Matthijs Elenbaas and Claes H. de Vreese, “The Effects of Strategic News on Political Cynicism and Vote Choice among Young Voters,” Journal of Communication 58, no. 3 (September 2008): 550–67, https://doi.org/10.1111/j.1460-2466.2008.00399.x.

17

Izabela Kapsa, “Political Trust vs. Turnout in Modern Democracies,” Polish Political Science Yearbook 49, no. 3 (September 30, 2020): 151–60, https://doi.org/10.15804/ppsy2020309,151-152.

18

“Daily Average Time Spent on Various Activities by Age Group and Sex, 2015,” Daily average time spent on various activities by age group and sex, 2015, inactive, April 3, 2019, https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=4510001401.

19

Canada: Top 10 Insights (Toronto: Edelman, 2021), https://www.edelman.ca/sites/g/files/aatuss376/files/trust-barometer/Top%2010%20Canadian%20Trust%20Insights.pdf.

20

2023 Edelman Trust Barometer Top 10 (Toronto: Edelman, 2023), https://www.edelman.ca/sites/g/files/aatuss376/files/2023-03/Trust2023_EN_Top10_Final.pdf.

21

Leticia Rodríguez-Fernández, “Disinformation and Organisational Communication: A Study of the Impact of Fake News,” Revista Latina de Comunicación Social, November 4, 2019, 1714–28, https://doi.org/10.4185/rlcs-2019-1406en, 1714.

22

Ibid, 1721.

23

Ibid, 1722.

24

Sinan Aral, “Truth, Disrupted,” Harvard Business Review, February 24, 2023, https://hbr.org/2018/07/truth-disrupted.

25

Maria Cristina Arcuri, Gino Gandolfi, and Ivan Russo, “Does Fake News Impact Stock Returns? Evidence from US and EU Stock Markets,” Journal of Economics and Business, May 2023, https://doi.org/10.1016/j.jeconbus.2023.106130, 8.

26

Dan Milmo, “UK Engineering Firm Arup Falls Victim to £20m Deepfake Scam,” The Guardian, May 17, 2024, https://www.theguardian.com/technology/article/2024/may/17/uk-engineering-arup-deepfake-scam-hong-kong-ai-video.

27

Nick Robins-Early, “CEO of World’s Biggest Ad Firm Targeted by Deepfake Scam,” The Guardian, May 10, 2024, https://www.theguardian.com/technology/article/2024/may/10/ceo-wpp-deepfake-scam.

28

“The Future of Cons and Frauds,” Chartered Professional Accountants Canada, January 5, 2024, https://www.cpacanada.ca/news/pivot-magazine/fraud-2024.

29

“AI Deepfakes Increasing Fraud Risks for Businesses, KPMG Survey Finds,” KPMG, March 11, 2024, https://kpmg.com/ca/en/home/media/press-releases/2024/03/deepfakes-pose-major-fraud-risks-to-canadian-businesses.html.

30

Ibid.

31

Sarah Repucci, “Democracy Is Good for Business,” Freedom House, August 3, 2015, https://freedomhouse.org/article/democracy-good-business;
Rebecca Henderson, “Business Can’t Take Democracy for Granted,” Harvard Business Review, January 8, 2021, https://hbr.org/2021/01/business-cant-take-democracy-for-granted.

32

Sarah Repucci, “Democracy Is Good for Business,” Freedom House, August 3, 2015, https://freedomhouse.org/article/democracy-good-business.

33

Daron Acemoglu et al., “Democracy Does Cause Growth,” Journal of Political Economy 127, no. 1 (2019): 47–100, https://economics.mit.edu/sites/default/files/publications/Democracy%20Does%20Cause%20Growth.pdf.

34

Carl Henrik Knutsen, rep., Democracy, Autocracy and Economic Development (V-Dem Institute, 2019), https://v-dem.net/media/publications/v-dem_policybrief_20_2019_v2.pdf.

35

Martin Wolf, “For All Its Faults, Democracy Is Still Better than Autocracy,” Financial Times, April 2, 2024, https://www.ft.com/content/9285ed6e-fb71-4b10-bf5c-b4f83a140675.

36

Monica Bulger and Patrick Davison, “The Promises, Challenges and Futures of Media Literacy,” Journal of Media Literacy Education 10, no. 1 (2018), https://digitalcommons.uri.edu/cgi/viewcontent.cgi?article=1365&context=jmle;
Alisha Reed Anderson, “Teaching Critical Reading: Media Literacy in the High School Classroom” (dissertation, University of South Carolina Scholar Commons, 2019), https://scholarcommons.sc.edu/etd/5192/;
Eun-mee Kim and Soeun Yang, “Internet Literacy and Digital Natives’ Civic Engagement: Internet Skill Literacy or Internet Information Literacy?,” Journal of Youth Studies 19, no. 4 (September 23, 2015): 438–56, https://www.dhi.ac.uk/san/waysofbeing/data/citizenship-robson-kim-2016.pdf.

37

James Weinberg, “Civic Education as an Antidote to Inequalities in Political Participation? New Evidence from English Secondary Education,” British Politics 17, no. 2 (July 3, 2021): 185–209, https://doi.org/10.1057/s41293-021-00186-4.

38

Fernando Feitosa, “Does Civic Education Foster Civic Duty? A Systematic Cross-Country Analysis of the Effect of Three Forms of Civic Education on the Sense of Civic Duty to Vote,” Canadian Journal of Political Science 53, no. 4 (October 6, 2020): 887–901, https://doi.org/10.1017/s0008423920000669.

39

Kara Brisson-Boivin, “To Address Online Harms, Canada Must Commit to Digital Media Literacy,” The Hill Times, May 1, 2023, https://www.hilltimes.com/story/2023/05/03/to-address-online-harms-canada-must-commit-to-digital-media-literacy/385274/.

40

Melissa Eddy, “A New Place to Learn Civics: The Workplace,” The New York Times, October 29, 2023, https://www.nytimes.com/2023/10/29/world/europe/businesses-civics-education.html.

41

Ibid.

42

Ibid.