TikTok Labels AI-Generated Content From OpenAI And Elsewhere

(CTN News) – TikTok made the news on Thursday that it would begin deploying a digital watermark known as Content Credentials in order to identify images and movies that have been submitted to its video-sharing platform and that were created with the assistance of artificial intelligence.

Also included in the release was the fact that TikTok will begin using this watermark.

Concerns have been raised by the researchers over the possibility that information created by artificial intelligence (AI) could be utilized to exert influence on the elections that will take place in the United States this next fall. Twenty different internet businesses came together at the beginning of this year to sign a contract that committed them to addressing this threat. The company TikTok was one of those businesses.

In accordance with the most recent way, a label will be affixed to movies and images that were produced outside of the domain of the service. At the moment, the firm is in the process of assigning labels to material that was produced by artificial intelligence within the app itself.

“We also have policies that prohibit realistic AI that is not labelled,” said Adam Presser, who is the head of operations, trust, and safety at TikTok.

“Therefore, if realistic AI (generated contents) appears on the platform, then we will remove it as a violation of our community guidelines.”

Presser is a member of the TikTok team.

This statement was made by Presser as he was being interviewed. The Content Credentials technology was developed by the Coalition for Content Provenance and Authenticity, which was created by Adobe (ADBE.O), Microsoft (MSFT.O), and other firms.

This coalition was responsible for the creation of the technology. Now, other businesses are able to make use of this technology without restriction.

The group that was responsible for the development of ChatGPT, OpenAI, is one of the organizations that has already committed to using it.

Meta Platforms (META.O), the company that owns Instagram and Facebook, and Alphabet’s (GOOGL.O) Google, which controls YouTube, have both declared their willingness to use Content Credentials. Meta Platforms is a subsidiary of Alphabet. One of Alphabet’s subsidiaries is called Meta Platforms.

In order for the system to operate effectively, it is necessary for the platform that is in charge of the distribution of the content and the designer of the generative AI tool that is in charge of its creation to come to an agreement on the standard that is used in the industry.

When an individual generates an image by utilizing the Dall-E tool that is made available by OpenAI, for example, OpenAI adds data to the file that can be utilized in the future to assess whether or not the image has been modified. In addition to that, OpenAI will embellish the final output with a watermark.

In the event that the image that was highlighted is later uploaded to TikTok, it will be automatically identified as having been produced by artificial intelligence.

The Chinese company ByteDance is the proprietor of the social media network TikTok, which has a total of 170 million members. Because of the recent establishment of a regulation in the United States, ByteDance is now compelled to sell TikTok or else it runs the risk of being barred from the market.

In an effort to prevent the statute from being interpreted as a violation of the First Amendment, both TikTok and ByteDance have filed a lawsuit against the government.

SEE ALSO:

Investments In Arm’s Share Price Falls After Disappointing Revenue Forecasts