It is often the marginalized who suffer most from the unintended consequences of new technologies. For example, algorithms that automatically decide who gets to see what or how to interpret images are affected by race and gender biases. People with multiple marginalised identities, such as blacks and people with disabilities, are at greater risk than people with a single marginalised identity.
That’s why I cringe when Mark Zuckerberg lays out his vision for the metasurverse — a virtual environment where many people can interact with each other — and says it will touch every product the company makes. As a researcher who studies the intersection of race, technology and democracy — and as a black woman — I think it’s important to carefully consider the values that are being encoded into the next generation of the Internet.
Problems have surfaced. Avatars, graphic characters that people can create or buy to represent themselves in virtual environments, priced differently depending on the avatar’s perceived race, racist and sexist harassment is emerging in today’s pre-immersive environments of virtual worlds.
Ensuring that the next iteration of the Internet is inclusive and accessible to all will require people from marginalized communities to take the lead in shaping it. It also requires strict regulation to hold big tech companies accountable to the public interest. Without that, the metasomes could inherit, if not worsen, the problems of today’s social media.
Utopian vision and grim reality
Early utopian visions of the Internet often assumed that life online would be very different from life in the real world. For example, people envision the Internet as a way to escape parts of their identity, such as racial, gender and class differences. In fact, the Internet is far from race-free.
While technological utopias convey desirable visions of the future, the reality of new technologies often fails to fulfill these visions. In fact, the Internet has brought new forms of harm to society, such as bias in the algorithms that automatically spread propaganda on social media and shape your online experience.
Mr Zuckerberg described the metasverse as a more immersive, embodied Internet that would “unlock a lot of amazing new experiences”. This is not just a vision for the future of the Internet, but a vision for the future of lifestyle. However off-target that vision may be, the metasomes are likely — like earlier versions of the Internet and social media — to have wide-ranging effects that will change the way people socialize, travel, learn, work and play.
The question is, are these consequences the same for everyone? History suggests the answer is no.
Technology is never neutral
Widely used technologies often have white male identity and body as the default Settings. Joy Buolomwini, a computer scientist at M.I.T., showed that facial-recognition software performed worse for women, especially those with darker faces. Other studies confirm this.
Even in the absence of race as a category for machine learning algorithms, whiteness is the default setting for these techniques. Unfortunately, racism and technology often go hand in hand. In the 2020 election cycle, black female politicians and journalists were disproportionately targeted with abusive or questionable tweets, and black and Latino voters were targeted by online misinformation campaigns.
This historical relationship between race and technology makes me worry about the meta-universe. If the metasurverse is a materialized version of the Internet, as Zuckerberg describes it, does that mean that already marginalized people will experience new forms of harm?
Facebook and its relationship with black people
The general relationship between technology and racism is only part of the story. Meta has had a poor relationship with black users on its Facebook platform, especially black women.
In 2016, reporters at ProPublica found that advertisers on Facebook’s AD portal might exclude people who see their ads based on a user’s race or what Facebook calls “racial affinity.” This option has been met with a lot of opposition because Facebook does not ask its users about their ethnicity, meaning users are assigned a “racial affinity” based on how engaged they are on the platform, such as which pages and posts they like.
In other words, Facebook is essentially profiling its users based on their behavior and preferences on its platform, creating opportunities for advertisers to discriminate against people based on race. Facebook has since updated its AD targeting categories to no longer include “racial association.”
However, advertisers will still be able to target people based on their presumed race through ethnic agents, which use combinations of users’ interests to infer race. For example, if an N advertiser sees from Facebook data that you have expressed an interest in African American culture and the BET Awards, it can infer that you are black and target ads to the products you want to market to blacks.
To make matters worse, Facebook routinely removes comments from black women who oppose racism and sexism. It’s ironic that comments by black women about racism and sexism are being censored — colloquially “zatted” — because they ostensibly violate Facebook’s policy against hate speech. It’s part of a larger trend of black women being punished online for expressing concerns and demanding justice in the digital space.
According to a recent Washington Post report, Facebook knows its algorithms disproportionately harm black users, but chooses to do nothing about it.
A virtual world of democratic accountability
In an interview with Vishal Shah, vice president of Metaverse, NPR host Audie Cornish asked, “If you can’t handle comments on Instagram, how can you handle t-shirts with hate speech on Metaverse? ? How do you deal with a hate rally in a virtual world?” Similarly, if black people are punished for speaking out against racism and sexism online, how can they do so in the virtual world?
Ensuring that virtual worlds are inclusive and promote democratic values rather than threatening them requires design justice and social media regulation.
Design justice places those without power in society at the centre of the design process to avoid perpetuating existing inequalities. This also means starting by thinking about the values and principles that guide design.
Federal law protects social media companies from liability for the posts and actions of users on their platforms. That means they have the right but not the responsibility to police their sites. Regulating big tech is critical to addressing today’s social media problems, at least until they build and control the next generation of the Internet.
The meta-universe and me
I have nothing against the metadverse. I stand for a democratic and responsible virtual world. To do this, however, I assert that there needs to be a better regulatory framework for Internet companies and a more impartial design process so that technology does not continue to be associated with racism.
For now, the benefits of the metadverse don’t outweigh its costs to me. But it need not stay that way.