TikTok Will Automatically Authenticate AI-Generated Content

Reuters reported that starting May 9, TikTok will use "Content Credentials" technology, a digital watermark that will indicate how images are created and edited. This feature will help the social networking platform automatically identify products created by AI.

Adobe was a pioneer in using this technology and it is increasingly being adopted by many companies, including OpenAI - the company that created ChatGPT.

TikTok's New Update Aims at AI-Generated Content

Researchers have previously raised concerns that unverified AI-generated content could be used to spread misinformation, affecting many elections around the world , including the US presidential election later this year. Other tech companies such as Google and Meta have also announced plans to use this content verification feature.

TikTok has been proactively labeling AI-generated content when users use its built-in tools. The latest update will apply to content created by parties not managed by TikTok.

For this feature to work, platforms like TikTok and AI content creation tools must coordinate and adhere to industry standards. If a person uses Dall-E, OpenAI’s image-generating AI tool, OpenAI will attach a valid label to the generated image. If that image is uploaded to TikTok, it will automatically be recognized as AI-generated content.

“Our users and creators are excited about AI and what it can do, and we want to make sure people have the ability to tell what is fact and what is fiction,” said Adam Presser, TikTok’s head of operations, trust and safety, in a statement to ABC News on May 9.

According to a statement posted on TikTok’s website in September 2023, the company asked users to proactively label AI products. TikTok was also testing a self-authentication feature at the time.

Source: https://thanhnien.vn/tiktok-se-tu-dong-xac-thuc-noi-dung-do-ai-tao-ra-185240509200106262.htm