Algorithmic Media Observatory

digital methods for internet policy and politics

Opening Remarks to the Senate Committee on Transport and Communications on the Online Streaming Act

by

Introduction

I am an Associate Professor in Information and Communication Technology Policy at Concordia University. My research addresses the intersection of algorithms and AI with media policy. I submit these comments in a professional capacity representing my views alone.

Background

I am speaking from the unceded Indigenous lands of Tiohtià:ke/Montréal. The Kanien’kehá:ka Nation is recognized as the custodians of the lands and waters on which we gather today.

That acknowledgment has great urgency. As a participant in the National Culture Summit, I witnessed a consensus to reimagine cultural policy post-Massey. A future where investments in the Arts and Culture are something other than a settler-colonial project, a future where Cultural Policy is key to governments’ responsibilities and their relations to the many peoples who find their home in Canada.

That future is also bound to public service media in Canada. For all the talk today about disinformation and threats to democracy, the academic evidence is clear. Canada’s resilience depends on public service media. Public service media requires funding and innovation. Change that cannot wait. Every year I am reminded how irrelevant public service media is becoming as I ask my students their own media habits.

These students are part of a new generation of cultural creators working on new platforms under precarious working conditions. We need only be reminded about the coordinated harassment against Clara Sorrenti most recently to know that who gets considered a cultural worker and protected as a cultural worker. I know this better thanks to Margaret McDonald and Dr. Valerie Webber’s research on sex workers online and my collaborations with Saskia Kowalchuk on the meme makers on Tik Tok. These creators need rights.

These are issues of great importance but not directly addressed, or at least clearly addressed, by C-11. I recommend that C-11 is narrowed in scope, and that subsequent legislation addresses the objectives of Canadian cultural policy, renews public service media, and develops rights and protections for digital creators.

C-11

The Act today has one clear objective: ensure that the CRTC has the capacity to regulate large, economically powerful domestic and international firms involved in broadcasting distribution. Online streaming regrettably, remains a broad term, but the mission-critical function of the new Act must address the convergence of large online video-on-demand services and traditional broadcasting distribution undertakings.

The maturation of the streaming service to a few dominant players indicates that online services have become cable by other means. C-11 must ensure the CRTC has the jurisdiction to regulate this converged broadcasting system.

To do so, I see 4 areas for improvement.

1.    Discoverability

2.    Privacy

3.    Reforms to the CRTC

4.    Scope

Discoverability

First, the lack of a clear definition of discoverability – as opposed to prominence or catalog quota – undermines a larger movement for algorithmic and AI accountability. I recommend the Act defines discoverability its consequences for regulated entities and its feasibility under the government’s commitments to the USMCA. In addition, I recommend a more coordinated effort across government to implement an Algorithmic and AI Accountability Act.

Privacy

The discoverability question is a proxy question that might be rephrased: does being a private data collector grant broadcasting distributors unfair market power? I think so.

As I have written with my co-author Bram Abramson, we recommend the act adds the following policy objective: “contribute to the protection of the privacy of persons and recognize the public and cultural significance of information about them.”

The amendment recognizes both the need to harmonize privacy and broadcasting policy, but also recognizes the significance of cultural data as a public commons that require care and protection, especially in an age of DALL-E and Stable Diffusion.

CRTC

Given the complexity of these matters, I am discouraged that the Act has not followed the BTLR report’s recommendations for CRTC reform especially the establishment of a Public Interest Committee. As someone who cares and participates actively at the CRTC, I worry the expectations that C-11 places on the CRTC are unrealistic, gives too much voice to corporate interests and too little to public interests. I recommend the BTLR recommendations, Section 1, be considered as part of C-11 or another Act in tandem.

Scope

Finally, I continue to not know the scope of the Act. The lack of clarity and unwillingness to narrow its application to Canada’s media system remains C-11’s enduring shortcomings. I recommend that Section 4.1 be clarified as well as how platform affiliates would be subject to regulation.

These matters are for today. What matters to me most are the future of cultural policy I discussed at the start that remain stalled due to the lack of reforms to C-11.

Trudeau loses meme war; Singh lone candidate with positive coverage

by

Analyzing the top seven Facebook groups that seemed organic and not third-party advertisers, we found that meme groups mostly engaged in negative campaigning, fostering group affinities and trying to vote out Trudeau.

Over the 40-day campaign, these groups posted 580 memes or on average 14.5 memes per day. As seen in Figure 1, NDP and anti-Trudeau Facebook groups were the most active.

Figure 1 – Number of memes posted by group type

We found:

  1. No active meme groups were dedicated to promoting Trudeau, Scheer, or May.
  2. Anti-Trudeau meme groups encouraged followers to vote Conservative for Scheer, though not as much as criticizing Trudeau.
  3. Singh was the most popular candidate in memes with NDP-leaning meme groups engaging in positive campaigning around their leader and party.

TrudeauScheerSinghMayBlanchetBernier
Negative2128711111720
Neutral368479521566563558
Positive01347302

Figure 2 summarizes the table above. Tone has been coded by one coder at present, so inter-coder reliability is not available.

Figure 2 – Tone in memes

From our first-hand experiences of collecting memes, we noticed that:

  1. Misinformation about Trudeau, especially allegations of sexual misconduct, had a receptive audience in anti-Trudeau groups, a popular theme throughout the campaign.
  2. Styles differed by partisan groups with NDP and Liberal groups seeming the most adept in referencing Internet culture where Anti-Trudeau memes created their own ‘internet ugly’ images shared by supporters.

After the election, we will be analyzing the content and partisan styles of memes as well as further code memes by issue so please keep following for more analysis.

Memes after Trudeau’s Brownface Incident

by

Highlights

Internet users were quick to ‘meme’ last week’s scandalous images of Trudeau dressed in ‘brownface’ and ‘blackface’. Memes differed by the political allegiance of those who shared them, illustrating how party supporters make memes to help their party’s public image.

Anti-Trudeau Conservative memes highlighted Trudeau’s hypocrisy, questioned his character, and characterized Liberal party accusations of racism as disingenuous.

Memes from the left focused on the broader social issue, choosing to address racism in Canada and the skin tone of its NDP leader Jagmeet Singh rather than targeting the specific images of Liberal candidate Justin Trudeau.

Pro-Trudeau Liberal memes sought to minimize the incident, focusing more on characterized Conservative leader Andrew Scheer as the real threat.

Analysis

The Algorithmic Media Observatory is tracking October 2019’s Federal Election in Canada specifically through memes, or images inspired by Internet culture.

Ugly and home-made but also incisive commentary and artful craft, memes offer a unique vantage point from which to observe public opinion and, crucially, how party supporters and engaged citizens use the internet to identify with parties, leaders and movements.

To date, we have identified roughly 30 Facebook groups posting memes about Canadian politics, some professional other seemingly user-generated. We have found that:

  1. There is a strong presence of anti-Trudeau and Conservative meme groups on Facebook. These groups are among the largest, most active in the sample.
  2. There are smaller, but equally active, meme communities for Liberal and NDP supporters that share memes daily. 
  3. Meme groups care about different issues (climate change in the left and center vs. political corruption on the right and center), but the Trudeau scandal mobilized all sides.

These three features are clearly exhibited in the top 20 Facebook groups by follower count.

Most groups post negative content about Liberal candidate Justin Trudeau. Anti-Trudeau groups have a much larger follower count (approximately 900,000 anti-Trudeau followers), although we hesitate to give much weight to follower counts given how easily these numbers can be artificially manipulated. In the past few days, some Liberal groups have remained committed to the party, but tried to distance themselves from Trudeau.

Consensus coded by Fenwick McKelvey and research team

Conservative and Anti-Trudeau Memes

Among more conservative circles, memes emphasized the hypocrisy of Justin Trudeau and his followers. Memes, such as Figure 1, use the scandal as a chance to discredit Liberals as hypocrites. In doing so, these memes re-frame a discussion about race in politics as an opportunity to discredit any accusations of racism in the campaign, especially from Liberal candidates and supporters. These memes suggest some blowback to the Liberal’s strategy of opposition research and negative campaigning  against the Conservative party. Anti-Trudeau memes, such as Figure 2, also connect Trudeau’s poor judgement in 2001 with tropes and opinions of his other moral shortcomings (like being too ‘gropy’ with women).

Figure 1
Figure 2

Left and NDP Memes

Unlike conservative groups, NDP supporters heavily mocked the prime minister’s actions as racist, as seen in Figures 3 and 4.

Figure 3
Figure 4

Emblematic of the NDP’s focus on social issues, the memes showed direct concern of racism, pushing themselves onto a socially moral high ground with their own leader, Jagmeet Singh. Memes questioned Liberal sentiment, presenting the defense of Trudeau as humorous and ridiculous as well as the calling out the ignorance of Liberals regarding structural racism in Canada.

Liberal, pro-Trudeau Memes

Pro-Trudeau, Liberal memes have attempted to do damage control by attacking conservatives. Their memes ignore the black/brownface actions and focus instead on discussing racism within Canadian politics in general. These memes identify the Conservative candidates, party and ideology as a worse version of Trudeau’s actions, implying that Trudeau is not being “as” racist as the Conservative party, as seen in Figure 5

We have also observed Liberal groups sharing a photo of Stephen Harper during an honorary Indigenous ceremony, removing the context of the image to fabricate their message. This strategy backfired, as Robert Jago explains:

Here’s a public service berating: Liberals, when you send this picture around and say – look Harper did it too – he’s not at a Halloween party, he’s at a Native ceremony. That paint, the headdress, you’re laughing at a Native ceremony. YOU are the racist asshole in this scenario – Robert Jago

Other Liberal memes also downplay the damage this will have to the Liberal party’s election fortunes, questioning whether the past transgression of Trudeau would lead to “minorities” to vote for Scheer as seen in Figure 6.

Figure 5
Figure 6

Through memes, we have a new vantage point to understand partisanship in Canada as well as the ways message control may operate on so-called ‘user-generated’ content.

As the black/brownface controversy and election generally progress, memes will continue to further dialogue and opinion on candidates and the election. Our analysis of this event offers an important snapshot of how different segments of Canada’s political spectrum interpret racist acts and highlights how politics and humour intersect.

October 9 Event: Alexia Maddox

by

Bitcoin Blockchains on Twitter
Alexia Maddox
October 9, 2018
4:30–6pm

Cryptocurrencies represent emerging financial technologies engendered through overlapping community values of decentralised peer-to-peer exchange, encryption technologies and an overarching agenda towards the disruption of centralised banking within the fiat economy. This paper will trace the development and shifts in public discourse within social media surrounding cryptocurrencies. The last five years have seen cryptocurrencies move from technological emergence to a broadening range of applications and history potholed with disputes, divergence, hacks and scams within the community. The accompanying influence of speculation has shifted the focus from social adoption to value volatility and seen the incorporation of associated technologies within banking and other organisational processes. The emphasis within public discourse has also followed a shift from bitcoin to blockchain. The study is grounded through a Twitter analysis of cryptocurrency-related social media discourse within the Australian context. The social media analysis works with social media archives of the Australian Twittersphere captured between early 2012 to May 2017. Access to this curated archive is through TrISMA and the timeframe under analysis aligns with the most detailed available dataset. The analysis seeks to characterise the emergence of public dialogue surrounding cryptocurrency use and application over time, focusing on peak engagement events. The key concepts directing the focus and interpretation of the social media analysis include financial inclusion, socio-technical disruption and social change. The whimsical quest of the study is to learn where the digital frontier has shifted to within this community and point to possible future developments. From a community studies perspective the case study represents an initial foray into data analytics to explore whether it is possible to detect the shifting shape and form of digital community through its environmental imprint (Maddox 2016). This methodological aspect of the work speaks to an attempt to generate a data recognition practice that can be deployed to search for signatures of social disruption within digital trace data.

Milieux Institute EV 11.705
Concordia University
1515 St. Catherine St. W

Faking It: Deepfake Porn and Moral Panic

by

Essay by Maggie MacDonald

2018 has been the year of deepfakes: algorithmically generated videos that insert faces into existing video footage, fabricating convincing scenes which never took place in reality. They’re named after the Reddit user who circulated many of the earliest videos made using the technique—one that can now be mastered by anyone with a computer and some free time. The resulting videos are convincing, in fact they’re nearly indistinguishable from real footage: politicians delivering fake speeches, for example, and endless Nicolas Cage appearances. But, perhaps unsurprisingly, internet denizens have put deepfakes almost entirely to use in the pursuit of manufacturing pornography. Porn consumers have now become the producers of their own scenes, swapping celebrities’ faces into thousands of preexisting porn videos.

Deepfakes are being received as the harbingers of a moral apocalypse, with news coverage proclaiming them “a looming crisis for national security, democracy and privacy” and warning vaguely but alarmingly, that “AI-assisted fake porn is here and we’re all fucked.” But this framing forgets a crucial point: the techniques and technologies behind this practice are neutral. Machine learning is a tool that can be employed for good or for nefarious purposes. More than that, deepfakes are the most recent on a continuum of doctored sexual material spanning back over a century. From Tijuana Bibles to slash fan fiction to revenge porn, pornography has long before now been produced within a culture that does not value the consent of its subjects.

It’s not that the moral panic levied at the practice is undeserved. It does, of course, have consequences that must not be minimized. But framing deepfakes as the exception rather than the rule detracts from our ability to engage with our culture’s larger, underlying devaluation of women’s consent and disrespect for their bodily autonomy. This problem, in essence, is not unique to deepfakes, or even restricted to pornography; it saturates our social and political life, especially women’s involvement in it. Deepfakes are ethically troubling to be sure, but looking beyond the panic we could perhaps observe these media makers to help understand emerging media ecosystems.

Only this time last year the technique required dedicated teams equipped with training, tech expertise and advanced editing equipment, but since the early months of 2018 the videos have worked their way onto message boards and laptops the world over. Deepfakes train a deep learning algorithm to recognize the characteristics of a given face using an image bank. As explained by technology journalist Samantha Cole, after enough of this “training,” the assigned nodes arrange themselves to “correct” faces that appear in selected videos. Using a machine learning algorithm, a home computer, publicly available photos and some spare time, anyone can now fake a hyperrealistic video.

This type of content borrowing is typical of today’s participatory culture. In this media ecosystem, audiences construct their own culture through content appropriated from mass media, reshaping it to serve their interests. This circuit of simultaneous production and consumption depends heavily on the engagement of the average participant, the producer-consumer, or as Henry Jenkins names it, the ‘prosumer’.

As prosumers, deepfake creators are gesturing towards what they want to see in their media. When they appropriate and transform porn and celebrity in new hybrid forms of expression, they are harnessing the technique’s potential to demonstrate that anything can be made real. Unlike with studio-produced porn, deepfake communities act like fans. A free economy flourishes and engagement is incredibly high in many porn-sharing forums, especially ones that allow for creative freedom. We could have seen it coming: before this year, the immense popularity of porn GIFs suggested there was an active audience, eager to produce as well as to consume pornography. Porn scholar Helen Hester has defended these forums as exemplary spaces for participatory interaction, which is at odds with the image of porn browsers as “passive, thoughtless, and wholly receptive”. The tactics of deepfake prosumers could have been predicted, considering these fans are enmeshed in the modern world of porn – a system already riddled with piracy and an alarming disregard for consent.

Actresses including Gal Gadot, Emma Watson and Scarlett Johansson have been targeted by the practice but practically there isn’t much legal recourse for victims of a deepfake — a large part of why the practice took off so explosively. The ethical issues of consent and objectification have made it clear that a video itself does not need to be real in order for the personal damages it incurs to be. Defamation or copyright law may be a good place to start, but as one redditor put it, “You can’t effectively sue someone for exposing intimate details of your life when it’s not your life they’re exposing.” While sites like Reddit, Discord and PornHub have theoretically banned deepfakes under nonconsensual pornography clauses, the videos are emerging faster than they can be contained. There is currently no straightforward route for getting videos like these taken down, given their free circulation, anonymous creation and ability to perpetually re-emerge after they first appear.

It’s unquestionably alarming that porn is being faked without the consent of those depicted and there remains alarmingly little recourse for victims. Moderation of these videos is nightmarish, but neither legal nor regulatory responses have historically kept pace with any forms of porn making regardless of the type of technology involved. Treating this practice as an unprecedented moral problem brought about by technology ignores much larger cultural problems – a troubling red herring.

In this emerging media system, not only are media producers and consumers transformed from two separate categories into a shared pool of prosumers, but they interact with each other according to a new set of rules which none of us completely understands yet. A moral panic only obfuscates the cultural context from which these technologies emerge. Rather than misunderstanding emergent practices like this one, we should closely consider these techniques and communities in order to help us understand our media futures.

_______

Cole, Samantha. 2018. “We Are Truly Fucked: Everyone Is Making AI-Generated Fake Porn Now.” Motherboard. January 24, 2018.

Hester, Helen, Bethan Jones, and Sarah Taylor-Harman. “Giffing a Fuck: Non-Narrative Pleasures in Participatory Porn Cultures and Female Fandom.” Porn Studies 2, no. 4 (October 2, 2015): 356–66.

Jenkins, Henry. 2013. “Layers of Meaning: Fan Music Video and the Poetics of Poaching” in Textual Poachers: Television Fans and Participatory Culture. Routledge.

This piece was originally produced for Pause Button – an online publication about technology and culture published by the Milieux Institute at Concordia University in Montreal.

Maggie MacDonald is an MA candidate in Media Studies at Concordia University as well as the coordinator of the Media History Research Centre at Milieux Institute. Her research focuses on the transformation of pornography as a cultural industry through the platformization of pornographic content online.