When You Hear Some Feedback, Keep Going Take It Higher

There would be no “viral video,” as we know it today, without Janet Jackson. Without her “wardrobe malfunction,” at the hands of that [White] Man of the Woods¹, the founders of YouTube may never have thought to pivot their online dating service into the first major video sharing platform², in their simultaneous roles as Product Managers (PMs) of the technology. It is, perhaps, worth discussing the number of technologies we are familiar with today that are the result of failed dating services³ built, presumably, by men that could not get dates with women without a technological assist; the focus of this paper, however, is ethics in the product management of digital products (digital product management) that sit at the intersection of the arts and technology. Particular attention will be given to video-based products, asking the following questions: what is the influence federal and local law, parent companies, and product teams have on the content users create? In particular, what is the impact of “shadow banning” and the strategic defense of the Digital Millennium Copyright Act (DMCA)⁴ on Black creators?

A first question to address, though, is: what is product management? The long answer: if one were to ask five different PMs, one would get five different answers; product management as a function is less than one hundred years old, having been started at Procter & Gamble and which was once considered a Marketing function, responsible for the marketing and positioning of products, particularly physical ones, as well as responsible for their delivery to the market, until the early aughts. At this time, products began to become increasingly digital and, in response, product management began spinning into its own set of functions, under the guise of the Agile methodology (a response to the ’70s Waterfall methodology) of software development, which largely required PMs to act as Project Managers, responsible solely for the delivery of the product to the market, by way of managing deadlines for each other function’s delivery.⁵ Today, PMs function fairly similarly, the primary difference being that they are now also responsible for having decision-making expertise, not only helping other functions in completing delivery on time, but also assisting them in knowing what needs to, can be, and should be delivered. How this practically occurs is dependent on each specific organization. The short answer: practicing product management requires PMs to take in business needs and user needs, while relying upon the specialized expertise of their colleagues in other functions, to make informed decisions about what products and product features need to, can be, and should be delivered.

As our world and products become increasingly digital, so are PMs. So, what exactly should be considered a digital product? From an Internet of Things (IoT) perspective, the most inclusive definition would include all products that interact in the digital space, which means everything from Amazon’s Alexa to Epic Games’s Fortnite (the metaverse being conceptually related, but yet another topic for another day).⁶ For the purposes of this paper, digital products are software applications (apps), particularly mobile apps, rather than web apps (regardless of whether the app creating entity supports mobile and desktop versions of the same app). This explicit area is of import because, in an era of Internet ubiquity, access to the Internet is not nearly as ubiquitous, with low socio-economic status (low SES) individuals, especially Black people, being significantly more likely to access the Internet via smartphone than desktop or at-home broadband, compared to their white counterparts who see relative parity across all three types of access / behaviors.⁷ Refining the scope of discussion further, this paper will address video-sharing apps focused on video-based content creation and their parent companies to examine the relationship between ethics, law, and digital product management of arts-centered apps. With all respect to 2007-era YouTube and though much of the same content lands on each of the apps that still exists, YouTube inclusive, the discussion will focus on the following apps as more apparently contemporary than YouTube: Vine (Twitter); Triller; Musical.ly / TikTok (ByteDance); and Reels (Instagram).

Vine, may it rest in peace, was founded in 2012, acquired by Twitter before its early 2013 launch and shut down in late 2016; the hook of Vine was that videos could be no more than six seconds and played on a loop and, somewhat ironically, one of the factors leading to its closing was that marketing professionals were moving to longer short-form video platforms, back to YouTube and forth to Instagram and Snapchat. (You may be saying to yourself, “Yeah. What about Snapchat? I love Ghostface Chillah!” Similarly to YouTube, though, its initial intent was video-sharing alone, not content creation, which can be seen even more explicitly in the fact that its videos were not meant to be shared broadly or even stored; for that reason, this paper will elide over its place in the market.) All the same, Vine’s great artistic constraint birthed its own era of influencers that were both content creators and editors; it should be noted that Black male content creator Andrew Bachelor (KingBach) was the most followed on the app.

Per Ryan Murphy, “When a new Supreme rises, the old one fades away.” Apps are not an exception to the rule, so it’s perhaps unsurprising that Triller and Musical.ly had already been around over a year when Vine closed down its operations. TikTok had also entered the market just before Vine left it and was a more direct competitor to it and, even more to the point, continues to be one to Triller (though, notably, not Musical.ly, which TikTok acquired in 2017, and merged into TikTok in early 2018). The intent of these apps was to allow users to upload short videos of themselves lip syncing to popular songs. For the sake of simplification I will focus this discussion on TikTok as a singular app, but I cannot stress enough that it is but one app in a sea of apps that perform similarly and this family of apps has its roots in YouTube, which was built in response to a pop culture moment centered on the body of and image of a Black woman that was proliferated through the media without her consent (nor ownership, which is important insofar as we might address the commercialization and commodification of Black bodies, both literal and of content, without the subjects and creators of that content receiving compensation).

One such feature, belonging to an app in the above-mentioned sea and that is worth mentioning briefly, is Reels, which was built to try to directly compete with TikTok, around the time the former 45th President of the United States was threatening to ban TikTok⁸ in the country, seemingly for being a Chinese-built app that was gaining popularity during the COVID-19 pandemic, which originated in China. The racist overtone here should not be ignored because it is to the point that the parent company of Reels is Instagram, an app known for allowing the spread of misinformation⁹ and for suppressing the content of Black creators via “shadow banning.”¹⁰ While many tech organizations would say that shadow banning does not exist or is the wrong term,¹¹ it is colloquially correct and used to mean the intentional filtering out of certain content from users’ content feeds via various Artificial Intelligence (AI) implementations.

It’s been remarked by several technologists that implementations of AI, algorithms for the purposes of this paper, cannot themselves be biased, but rather take on the biases of the team(s) that build them.¹² Rather than re-hash this argument, the following will attempt to address the question that falls out from this truth: Again, what is the influence federal and local law, parent companies, and product teams have on the content users create?

Turning to the law, the green field for ethical digital product management of video-sharing apps focused on video-based content creation can begin to be found. The arts as intellectual property (IP) are largely covered by federal copyright and design patent law; the DMCA, as mentioned earlier, is an Act of particular consideration for those building and / or using such technologies, as it prohibits the use of technologies as a way to circumvent the country’s base copyright laws.¹³ Those base copyright laws, in addition to their protections for the IP itself, also provide guidance on fair use:

the fair use of a copyrighted work…for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include—

(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;

(2) the nature of the copyrighted work;

(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and

(4) the effect of the use upon the potential market for or value of the copyrighted work.

The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors.”¹⁴

While there are myriad other laws that regulate both tech and data, DMCA is the most integral to understanding the rising ethical challenges digital PMs are facing and what their opportunity is to have positive social impact. Currently, DMCA could be interpreted to be relatively restricting, compared to how social media and its users operate. TikTok, in its Community Guidelines, states:

We encourage everyone to create and share original content. Content that infringes someone else’s intellectual property rights is prohibited on our platform and will be removed if we become aware of it.”¹⁵

But isn’t TikTok in the business of profiting off of content built by users? Yes, and that content is required to be owned by the creator or made under fair use, but also becomes TikTok’s as part of their Terms of Service (ToS).¹⁶ Complicating things further, to scale the enforcement of their ToS, TikTok, like many companies, uses a violation detection algorithm to remove content that violates the ToS, including violations based on IP infringement.¹⁷ Surely there is a product team that monitors this algorithm and adjusts it, but it is still biased in favor of non-Black creators, like many apps’ algorithms, which becomes even more of a vicious cycle when this algorithm is tied to the product’s core value proposition. For TikTok this is its video recommendation algorithm.

Unlike other platforms which were built on the idea of following accounts of people you know (even if that’s following Jay Z’s Twitter, knowing he follows no one, not even Beyoncé), TikTok’s product is seeded on user data and uses that to curate a “For You Page” for its users and of which accounts followed by and accounts following are smaller factors in the algorithm.¹⁸ Because the app is also built around remixing of other users’ content, including both audio and video, what this means in the worst case scenario is that a trend deploying fair use of copyrighted media can rapidly come into the zeitgeist via memetic function (and just as quickly leave it) and the original content that produced the trend can be taken down for an alleged DMCA violation, while subsequent videos referencing it can gain more popularity, leading to the attribution of trends to the wrong individual(s). If the erasure of the original creator’s identity weren’t bad enough, they can be flagged and suppressed in the recommendation algorithm over time, shadow-banned, the more such “violations” occur.

Given that TikTok also profits off of relationships with content creators, insofar as taking them on as “sponsored” creators, paying them to make content that will help attract and retain users that drive their valuation, the above-mentioned side effects become even more pronounced, in that original creators of trends that deploy fair use of copyrighted media have reduced ability to financially compete in the “influencer” market and those to whom credit is wrongly attributed have an unfair advantage in that same market. Demographically, the original creators also tend to be Black, compared to their counterparts who tend to be non-Black. An example of this is the story of Jalaiah Harmon, creator of the “Renegade” dance trend on TikTok that catapulted Charli D’Amelio to fame.¹⁹

One ethical path forward for similar companies is to build out corporate governance policies that require products to be built within an ethical framework appropriate to the business, including regular ethical audits, and that lay out response plans to discoveries of unethical behaviors or operations that avoid common response mistakes. However, such policies are often not implemented at the point of an organization’s founding and are more costly and difficult to implement later on²⁰; building out separate programs (and appropriately managing any conflicts of interest) that educate legal professionals on how the built technology functions and that lobby for legal amendments to outdated laws that pose ethical risks to the users of that technology are worthwhile endeavors, which are maybe more achievable, as the law’s understanding of technology catches up to the present.

Education of legal professionals could look like easy-to-read product documentation²¹, published glossaries of technical terms relevant to the organization’s technology, or, borrowing from the legal profession, technical clinics where technology professionals advise lawyers on the details of technologies relevant to particular cases as consulting experts. Of course, there are any number of solutions that could work, but the goal should be for legal professionals to have an accurate understanding of the complexities of technologies relevant to their practice, even if the language of that understanding is abstracted away from the technical terms. Legal professionals, necessarily, should also include legislators; their dearth of understanding of technology is the reason why the law itself is behind,²² particularly in its understanding of how fair use is regulated internally to apps. To be clear, it is tech regulatory law that is more behind than copyright law, although certainly there is an argument for expanding the definition of fair use to consider more modern technologies.

A pessimist reading this is aware, as I am, that many companies will not engage in this work because it lacks financial incentive in a capitalist market. So, one might then turn to find hope in the people internal to a company that builds the product to operate in ways that are more ethical and considered. In her book, Continuous Discovery Habits, Teresa Torres says, “One area that product teams often overlook is ethical assumptions.”²³ This needs to stop; ethical assumptions should be the first thing addressed in product development, so as to make a green field for users, rather than for the business to profit. A proposed framework is to look at inclusion, diversity, equity, and accessibility as building blocks for product development. One model for understanding this is concentric circles, with accessibility as the outer circle and inclusion as the innermost circle; this framework recognizes that while there is a desire for the inner circle to be as big as possible, it is almost always the smallest.

The marriage of the Internet and the arts has created endless opportunity for discovery of both the self and others, but while the law continues to fail to make the Internet a safe and equitable space and companies continue to respond to monetary incentives, it is up to PMs (and product teams, broadly) to build ethically to create more ethical apps and a more ethical future.