JAKARTA - TikTok has long been the platform of choice to view interesting videos, but who would have thought the video application turned out to be a hotbed of misleading information.

Anyone using TikTok to learn about COVID-19, climate change or the Russian-Ukraine invasion is likely to find misleading information.

This is evident from a study conducted by NewsGuard, a company that often monitors misinformation.

The researchers claim nearly 20 percent of the sample search results for headlines featured misinformation. The false claims range from Russia's invasion of Ukraine to the COVID-19 vaccine and the January 6, 2021 attacks on the US Capitol building.

Mereka, para peneliti menemukan bahwa dengan metypekan pertanyaan yang tidak abahaya juga dapat menampilkan saran yang penuh dengan informasi salah.

For example, simply searching for the words "climate change" and TikTok will display the results of climate science refusal. In addition, they also found that nearly 1 in 5 videos were automatically suggested by the platform to contain misinformation.

The search for information on mRNA vaccines also produced five videos (from the first 10) containing misinformation, including baseless claims that a COVID-19 vaccine causes permanent damage to children's critical organs.

Steven Brill, founder of NewsGuard, said the amount of misinformation and the ease of finding it was very disturbing, considering TikTok is very popular among young people.

Brill also questioned whether TikTok's parent company, ByteDance made enough efforts to stop misinformation, or whether they deliberately allowed misinformation to proliferate there.

In response to this, to NBCBoston, Thursday, September 15, a TikTok spokesperson said they did not allow false and harmful information on their platform, and claimed to have removed it.

TikTok itself has been trying to remove nearly 350,000 videos of misinformation related to the 2020 US presidential election by the end of the year.

In addition, the company uses AI to filter videos, and withdraw clips that are automatically flagged or send them to human moderators.

However, this approach does not catch quite a number of accounts violating the rules, especially those avoiding the use of AI's easy-to-detect keywords.


The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)