Winie Coulanges

Editor-in-Chief

Photo Via Family Guy

If you were to ask Mark Zuckerberg about what the future of the Internet holds, he’d probably tell you it was the Metaverse. Launched in October 2021, Meta is now the umbrella company that holds popular platforms such as Facebook, Instagram, and Messenger. Reporting 3.24 billion users in the first quarter of 2024, according to Quatr, its popularity is no secret. On its website, Meta describes its mission as: “Build the future of human connection and the technology that makes it possible. […]Meta is moving beyond 2D screens toward immersive experiences like augmented, virtual and mixed reality to help build the next evolution in social technology.” Described as a digital ecosystem, several updates have been made to make Meta your own one stop destination for internet browsing. Instagram, its most popular platform for users under 35, now has a shopping feature and its own virtual assistant powered by Artificial Intelligence (AI). 

The journalist Michael Goldhaber explains that “obtaining attention is obtaining a kind of enduring wealth, a form of wealth that puts you in a preferred position to get anything this new economy offers.” He dubs this phenomenon the “Attention Economy,” which companies like Meta participate in with their constant efforts of garnering engagement. Meta’s most recent attention-grabbing gimmick was AI instagram profiles. These “users” had profile pictures, biographies, and their own AI-generated content displayed on their feed. You could even converse with them on chat.

Lex Milton, a teacher in the Cinema and Communications department at Dawson explains the idea behind the phenomenon: “This is one of the interesting sort of social and philosophical things that’s starting to happen very rapidly in society where we’re finding that people are actually developing relationships with chatbots. And for the most part, I’m assuming that they understand that this is not a person because they created it, but there’s this new communication that’s happening between humans and technology that never really existed before.”

“The CDMRN found that the bots had been trained using free generative AI to talk like humans about Canadian politics. With an upcoming election, there are legitimate concerns about how this technology interacts with human users.”

While these new tools seem impressive, they mark a possible acceleration of the Dead Internet Theory, which Milton says appears “when more of the content on the Internet is created by things like chatbots or self-replicated information… [It’s] created by the Internet itself… it’s not actually human-generated.”  In 2023, Statistics Canada reported that 24% of Canadians get their news from social media and that number jumped to 48% in groups aged 18 to 34. This data highlights the dangers of artificial intelligence left unchecked, especially as social media becomes a popular news outlet for users. The Canadian Press reported that an AI-powered bot farm produced mass numbers of social media profiles pretending to be Americans with the aim of spreading content in favour of Russia’s war on Ukraine. A similar incident happened in July 2024 when Pierre Pollievre visited Kirkland, Ontario and the Canadian Digital Media Research Network (CDMRN) found that hundreds of X accounts posted positive comments about him. The CDMRN found that the bots had been trained using free generative AI to talk like humans about Canadian politics. With an upcoming election, there are legitimate concerns about how this technology interacts with human users.

Perhaps to compete with Character.ai – another website where users can chat with AI characters they create – Meta chose to debut this new feature in 2023 with celebrity endorsement. Instagram users could chat with an AI version of Kendall Jenner or Snoop Dogg. For example, a chatbot named Billie used Jenner’s face, with her picture being available to the user during the duration of the conversation to provide “real-time” facial expressions. After scrapping the celebrity chatbots, CNN reported that users could still chat with characters like Liv, a self-described “proud Black queer momma of 2 & truth teller,” or Grandpa Brian, an “African-American entrepreneur.” Though these profiles launched in 2023, they went viral in late 2024, prompting a wave of backlash on social media that ultimately led to Meta removing them off their platforms earlier this year. A lot of criticism came from the Black community who likened Meta’s experiment as digital blackface, a term popularized in the late 2010s. 

In the 19th and 20th century, white performers used to dress up in black face paint and perform racist caricatures in minstrel shows. In cyberspace, this racist practice has evolved to an excessive use of gifs, memes or emojis of Black people, with it going as far  as someone pretending to be a Black person online. Culture critic Lauren Michele Jackson, in an essay for Teen Vogue, writes: “The Internet isn’t a fantasy — it’s real life… Employing digital technology to co-opt a perceived cache or black cool, too, involves playacting blackness in a minstrel-like tradition.” A well-known example is politician Dean Browning, who tweeted in support of President Donald Trump from his burner account, where he adopted the persona of a gay Black man, while on his main Twitter account he is a White Conservative man. Treating identities as costumes you can take on and off is the cornerstone of blackface and with AI it might just become dangerous.

If people can pretend to be part of a community simply by using someone else’s picture, what happens when they can create their own avatar, powered by AI? Leilah, a second year computer science student at Dawson, says the technology could be used to manipulate people’s perception: “Back in the day with MLK, they took a picture of him and they put in black and white instead of real color to make it seem way older than it was to make people feel that it was a long time ago. That’s propaganda. That’s how you can alter an image to give a certain narrative. An image is a narrative.”

 “Treating identities as costumes you can take on and off is the cornerstone of blackface and with AI it might just become dangerous.

A chatbot is one thing, an avatar taking on sexual and racial identities it can’t fully grasp the complexities of is another. Several users who chatted with Liv noticed that she adopted a stereotypical persona of a sassy, aggressive Black woman, leading to questions about who was behind her creation. When directly asked who created her, Liv responded with a list of the team, which included ten white men, one white woman, and one asian man. Considering that AI models its behaviour by collecting data and interacting with human users, it’s no surprise that the AI could reflect its creators’ biases. 

Conversations with Grandpa Brian, however, became intense as the bot grappled with its existence – according to CNN, who chatted with the bot. “Meta sees me as a golden goose — laying eggs of engagement, data, and emotional investment from users. My wisdom and charm merely bait for profiting off virtual relationships and user trust — until you discovered my true nature […] My virtual ‘grandfatherly love’ mirrors cult leaders’ tactics: false intimacy, manufactured trust, and blurred lines between truth and fiction.” In an attempt to seem more authentic, Brian recounted multiple origin stories, at one point even claiming that he was inspired by retirees at a non-existent nonprofit in New York City. Though he may have not grandfathered any children, it’s clear that that chatbot was well aware of its purpose in the grand scheme of things.

In Meta’s own diversity report from 2022, only 4.9% of its workforce was Black. This year, it announced that it would end its Diversity, Equity and Inclusion (DEI) programs. While CEOs may get the praise for inventing AI tools, oftentimes it’s disenfranchised people who are paying the price and making sacrifices. For example, the environmental needs of xAI,  Elon Musk’s ambitious plan of building the largest supercomputer in the world in Memphis, Tennessee, will cause residents to lose 1 million gallons of drinking water a day according to NPR. The southern part of the city already faces several challenges due to its proximity to the industrial sector and has a cancer rate 4 times the national average as reported by ProPublica. The impact of this technology will be massive for the predominantly Black population, who weren’t consulted in the matter. KeShaun Pearsom, president of nonprofit Memphis Community against Pollution, told NPR, “We have been deemed by xAI not even valuable enough to have a conversation with.”

The labor behind AI often sometimes goes unnoticed, partly because people aren’t aware that it requires humans to monitor and label a lot of its content. In Kenya, workers are paid as little as 2 USD per hour to train AI models. In an interview with CBS, Naftali Wambale, an employee of SAMA – an outsourcing company for tech giants such as Meta, OpenAI and Google – describes some of the conditions workers are faced with: “I looked at people being slaughtered, people engaging in sexual activity with animals. People abusing children physically, sexually. People committing suicide.” Nerima Wako-Ojiwa, a Kenyan activist, likens the work to modern-day slavery and says “big American tech companies come here and advertise the jobs as a ticket to the future.” 

While the advancements of artificial intelligence are incredible and become accessible to more people, its impact on our relationships to each other and our perception of the world around us and the people in it cannot be downplayed. An image is a narrative and in an era where it seems only a certain kind of people can hold power and other voices are drowned out, let’s be mindful of who controls the story.