Anonymous influencers are not a new phenomenon. Think of Marshmello basing his success entirely around his music in an iconic helmet or Corpse Husband continually charming the internet with his deep voice and veiny hands. Bygone is the era of a public figure’s influence centered solely around their physical existence. So what’s up with virtual, anime-inspired avatars popping up on the YouTube sphere lately? How do these motion-tracked personalities work? And why are YouTube stars like PewDiePie and Pokimane so eager to jump on the trend?

Virtual YouTubers (shortened VTubers) are online entertainers disguising their appearance using a customised digital avatar. These avatars are typically two-dimensional or three-dimensional creations that resemble anime characters with large eyes and boisterous personalities.

Animated with voices provided by the creator themselves, some of these avatars have complex and mythical backstories whereas others are presented as average creators who sing, dance and live-stream video games.

VTubing is a movement which took years to achieve the phenomenal status it has today. The community began back in 2011 with Ami Yamato, a virtual vlogger based in London who has a penchant for Starbucks and strolls around a seemingly-real world. Harbouring a mix of vlogs, movie parodies and sketches on her channel, the pioneer was quick to be followed up by a well-known character that is almost 60 years old: Barbie.

In 2015, the California-based toy company Mattel jumped on the concept of virtual vlogging as a marketing stunt to appeal to an increasing digital audience. Actively posting two vlogs a month, which takes over four weeks to produce, the teen vlogger tackles various cultural conversations and YouTube challenges on her channel.

Though virtual vlogging hatched into existence in 2011, the trend really took off with Kizuna AI’s debut in 2016. Credited with coining the term ‘VTuber’ herself, the virtual creator is hailed as the pioneer of the peculiar genre. Kizuna’s avatar plays on Japan’s famous anime tropes, featuring a bubbly girl donned head-to-toe in a white sailor’s uniform with giant bows and opera gloves. The VTuber runs two accounts—her main channel includes music covers, internet challenges and even a virtual bar where you can sit back and just hear Kizuna talk for minutes at a stretch. Her other account, AI Games, is dedicated to gameplay videos such as Fall Guys, Minecraft and Beat Saber.

Close to 3 million subscribers on her channels, Kizuna AI became the first VTuber to reach a global audience. Her large-scale impact even secured her the position as a cultural ambassador for the Japan National Tourism Organisation. Other virtual content creators were quick to witness a boost in following ever since Kizuna’s game-changing success. The OG VTuber ultimately paved the way for more creators specialising in all types of content ranging from house tours to mukbangs and ASMRs.

One thing to note before we dive into all of the technical aspects of the genre is that VTubers aren’t complete AI creations. There is a human behind their actions, controlling the avatar majorly through motion capture.

The process usually features creators fitted with motion trackers to record their movement and recreate various actions. These actions are then mapped over a shape and proportions of an animated character which is then rendered onto a background or live-streamed. This technology is what Kizuna operates on, thereby allowing the creator to interact with fans in real time at exhibitions, give interviews on live TV and even perform in concerts with the avatar at the forefront. VTubers are also responsible for voicing their own characters—although some of them use audio distortions to morph their speech in order to maintain complete anonymity.

But all of this sounds a bit labour-intensive and time-consuming, doesn’t it? While VTuber-dedicated agencies like Hololive Productions and Activ8 use Hollywood-grade equipment to crank out music videos, skits and game streams just about every day, smaller creators often rely on various animation and avatar creation software like Steam’s VRoid Studio, VTuber Maker and FaceRig to keep up with the fast-paced world of YouTube content.

“You don’t need to be artistic to use the program,” a top review on the Steam store reads. “You can use the tools provided to make accessories like horns, wings, tails, glasses and earrings.” Frequent users of these software back up the fact that “with enough practice and skill you can make top-notch creative models.”

WANNA SEE SOME TECH?!! This is Miko being mocapped live by technician with facial tracking! pic.twitter.com/a3A5gLWANv

— CodeMiko (@codemiko) November 29, 2020

After their initial resounding success in Japan, VTubers have found western integrations in the recent past. Canadian YouTuber Pokimane debuted a pastel-coloured 3D anime model of herself last year during one of her Twitch streams. PewDiePie is also among other famous YouTubers who have dipped their toes into the wonderful world of VTubing.

Global fans, when interviewed by the BBC, stated that the biggest contributor to the rise of virtual YouTuber is “the huge audience outside Japan who normally have interests in Japanese media and culture such as anime.” In the same interview, Takeshi Osaka, founder of the company behind Kizuna AI stressed that a VTuber’s pseudo-realistic presence is what makes them so appealing. “What separates VTubers from regular anime characters is that you can believe they actually exist,” he added.

In a report released by YouTube, the platform highlighted the fact that 47 per cent of its viewers are open to “watching content from creators or characters who are fictional or virtual.” “We saw VTubers start to take off right at the end of 2017…and it’s continued to grow ever since,” said Kevin Allocca, head of culture and trends at YouTube to the BBC. He further pointed to Kizuna AI as a major influence on the spike in VTubers’ global popularity.

VTubing essentially blurs the line between AI and reality, opening up a realm of endless possibilities for online creators. “No-cam streams are so comfy,” Pokimane said during her viral ‘VTube-worthy’ livestream. “Especially because it gets tiring to get people to stop commenting on you. I’d rather have people comment on this cute little anime drawing of me, you know?”

In the past, the YouTube star has been criticised for her ‘daring’ livestreams without any makeup—an expectation that pressurises streamers when it comes to their looks. Influencers on Twitch, especially women, are expected to show their faces and look good while streaming gameplays.

“I have noticed a decrease in viewership on days I don’t use a face cam, sometimes as much as 15 to 20 per cent,” admitted Twitch streamer Nikatine in an interview with Polygon. The streamer further listed out the fact that it’s not always feasible for a livestreamer to show up on-stream. “It’s challenging when there’s a heat wave to use the lights and makeup,” Nikatine continued. “There are plenty of days I just want to relax, not wear any makeup and just stream. When I don’t use a face cam, I get messages all the time in chat like, ‘why no cam, streamer?’ and ‘streamer, use cam’ and it makes me not want to do it even more.”

This is the perfect breeding ground for VTubers. A digital avatar proves useful for online personalities who want to stream when they’re not “feeling their best” or for a whole new generation of influencers who want to maintain their anonymity yet build an intimate relationship with their audience.

In less than three years, VTubers have essentially morphed from a subculture to big business. Kizuna AI can now be found in ads for instant cup noodles and eye drops. Dedicated VTuber talent agencies like Hololive—where VTubers undergo idol-level training via interviews and auditions—further give the trend some well-deserved PR. GREE, one of Japan’s biggest mobile app developers plans to invest $88 million over the next two years into developing virtual talents.

“We believe that human beings need avatars beyond nicknames and profile pictures,” said GREE spokesman Kensuke Sugiyama to the BBC. “Although virtual talent is currently a niche area of entertainment, we believe that attractive 3D avatar characters and their activities in virtual worlds will take people to the next stage of the internet.”

With #VTuber gathering over 614 thousand videos from over 21 thousand channels on YouTube, there is no doubt that these digital influencers will change the future of entertainment. They could change how brands market their products and how we interact with technology. And as we shift into a digital-first era, the concept of a VTuber is becoming increasingly relevant and feasible.

YouTube’s recommendation system is not set in stone. Every year, the company makes small changes to its ‘Up Next’ sidebar to reinforce its quicksand algorithm. In 2019, a research team from Google Brain, Google’s AI division, began testing a new algorithm that incorporated what is known as ‘reinforcement learning’ to build a “long-term addiction machine.” The new AI, called Reinforce, was designed to maximise user engagement over time by steering them into different parts of YouTube, eventually expanding their taste—instead of suggesting videos similar to the ones they had already watched.

Reinforce was YouTube’s most successful launch in two years. At an AI conference, Minmin Chen, a Google Brain researcher, admitted that sitewide views increased by nearly 1 per cent—which on YouTube’s scale surmounts to million more hours of watch time and revenue from advertising. She added that the new algorithm was already starting to alter users’ behaviour.

Enter YouTube’s far-right creators. Specialising in content targeting cross-genre exploration, these YouTubers tremendously benefitted from the algorithm changes, which unknowingly facilitate far-right radicalisation. They cleverly called out left-wing biases in videos reviewing the latest Star Wars episode and ranted about feminism while streaming Call of Duty—both moves attempting to ‘red-pill’ (an internet slang term for converting to far-right beliefs) movie buffs and gamers with niche content.

On a platform where 70 per cent of the user engagement is aided through recommendation, these entertainers perfected the art of tumbling viewers down a rabbit hole of far-right content. They built their audience with a subversive yet satirical take on leftist issues—fostering an addictive experience that shuts out all other views.

A probe by The New York Times proved the algorithm to be a useful recruiting tool for far-right extremist groups. Bellingcat, an investigative journalism website, found YouTube cited as the most frequent cause of members’ red-pilling in a far-right chat room. An analysis of around 30,000 Twitter accounts affiliated with the alt-right by VOX-Pol, a European research group, found accounts linked to YouTube more often than to any other site.

“They’re not selling politics as politics but conservatism as a lifestyle brand,” stated Ian Danskin, creator of Innuendo Studio, in a video titled ‘The Alt-Right Playbook: How to Radicalize a Normie’. He states that the practice of abandoning progressive principles and embracing conservatism is sold to its increasingly-pigeonholed viewers as “something that will make them happy.”

In his video, Danskin breaks down the steps taken by these creators into five main actions: identify the audience, establish a community, isolate, raise their power and give them a mission. With comments like “This was me two years ago,” “Can confirm this is how it works,” and “Damn dude, this hits hard,” Danskin makes his point as an active member of BreadTube—a collective of left-wing YouTubers united by their shared interest in combating the far-right to deradicalise viewers.

Often interchangeable with LeftTube, videos made by these crowdfunded creators mimic the aesthetics of right-wing YouTube by mixing politics with other mainstream interests like films, video games, popular culture and philosophy. Creators in this movement don’t get outraged by far-right ideals, but instead feature a theatrical yet didactical style to convey leftist thoughts, adopting a laid-back, rolling-my-eyes approach to counter far-right propaganda.

Initially created on Reddit, the term BreadTube comes from Peter Kropotkin’s The Conquest of Bread, a book on anarcho-communism. The community is highly decentralised, mostly running on mutual cameos and shout-outs between creators. Major figures associated with BreadTube are ContraPoints, Philosophy Tube, Lindsay Ellis and Hbomberguy. The label imposed on these creators, however, is highly-debated, with these YouTubers identifying to the term in varying degrees.

The core of BreadTube’s strategy is to hijack the YouTube algorithm to help burst its political bubble and foster a space for deradicalisation. BreadTubers use the same titles, tags and descriptions as far-right YouTubers so that their content is recommended to the same audience as the far-right. In some cases, these channels and creators respond directly under these far-right videos to increase their exposure and redirect traffic.

The success of BreadTubers can be quantified by the number of likes and comments amassed under their videos. The movement is referenced by academics as a case study in decentralisation. With contested claims of the YouTube algorithm increasingly promoting far-right ideals, BreadTube is definitely a step in the direction of beating the far-right at their own game. And quoting Danskin himself to sum up the success of the movement, “One thing we have that the alt-right doesn’t is hope.”