Bytedance, the mother company of Douyin and TikTok, became the first to use artificial intelligence to hook their users in an unprecedented way, says China internet expert Matthew Brennan to Play Crazy Game. The TikTok algorithms turns its users into addicts, in the same way drugs do, says Brennan.
Play Crazy Game:
According to Matthew Brennan, a technology expert and author of the book Attention Factory, Chinese ByteDance, creator of TikTok, knew very well what she was doing when she developed the app.
The author claims that TikTok uses one of the most sophisticated recommendation algorithms in the world and that its resounding success did not happen by chance. “Living in China, I saw firsthand the growth of Douyin (TikTok’s name in that country) in 2017, and the impact it had on everyone around me,” he says.
Behind the application is a highly efficient technological engine, capable of automatically auditing the millions of videos published, categorizing them one by one with keywords. First, each video is released to a few hundred active users for a kind of test. Then there is a crossing of information, the so-called metrics, which map the number of views, “likes”, comments, average viewing duration, shares, etc. All to identify the most popular content and send it to the next level, where it will be released to thousands of active users. The process repeats itself and, according to the result, the content continues to be sent to the next level, released to ever-larger audiences, reaching into the millions.
“ByteDance was the first Chinese internet company to fully dedicate itself to the then-new recommendation technology and to commit to the difficult task of creating a tool that challenges the status quo of human curation. The initial gamble paid off. The foundations of TikTok’s success were laid many years before the app itself was built, and it’s no coincidence that ByteDance was the company that created it,” says Brennan.
All the success of TikTok comes down to the recommendation tool, as it is what hooks the user to content that they like, giving them the false feeling of controlling what they see by moving their thumb up, triggering an infinite scroll bar, where you lose track of time.
By falling into recommendations, the user submits to what the application wants him to see. The more the tool gets it right, the more likely the user is to stay online, ingesting the little reward pills that, like any other drug, will become increasingly irresistible and uncontrollable.
To better understand how the chances of the tool getting it right are high, making the content addictive, we can make an analogy of how recommendations between humans are and how the tool works. Normally, when we read a book that we believe is of interest to someone, we recommend reading it, but buying the book, actually reading it and giving us feedback is a long process, which is 100% at the person’s discretion, and may even not work. in nothing.
The algorithm doesn’t work that way. Brennan explains that it has so-called “machine learning”, that is, it has the ability to learn by tracking user behavior. “What makes TikTok so addictive is that it learns what you like and what you don’t. And it does it very quickly because in one minute you can watch five or six videos. In that time, you have to discard or watch the video, revealing your preferences. In this way, ByteDance can get a lot of information in a very short time”, clarifies the author and adds: “It is an extreme customization”.
The way the algorithm works, the control that the user has over what he sees is practically non-existent. While the person thinks they are making their own choices, they do not realize that they are only providing information on the “substances” that should be put in their addictive reward pills.
Matthew Brennan is a speaker at the China Speakers Bureau. Do you need him at your (online) meeting or conference? Do get in touch or fill in our speakers’ request form.
Are you looking for more internet experts at the China Speakers Bureau? Do check out this list.