The platform is rolling out a new feature that enables them to reduce the amount of AI-made media appearing in their streams. This step comes as the platform disclosed that it presently contains in excess of 1 bn AI-generated clips.
Testing and Rollout
The new functionality is in a pilot phase over the following weeks before a full-scale rollout. This development comes amid a surge in AI content, spurred by emerging AI video creators.
The Broader AI Trend
Recent reports that approximately 10% of the quickest rising video streams on another major platform only show clips produced artificially. A significant portion of this material is classified as mass-produced AI material, a label for substandard, bulk-generated content that is frequently illogical or surreal.
“We know from our community that many people enjoy content produced via AI tools, from digital art to science explainers, and we want to give people the option to see more or less of that, depending on their own preferences,” said a spokesperson.
AI on the Platform
TikTok shared that there are currently 1.3 billion videos on its service tagged as artificially created. Overall, over 100 mn media items are added to the platform daily, making AI-generated material a relatively small portion of its total content library.
Customizing Your Feed
Members will be able to decrease or raise the volume of artificial media they watch by opening the content preference option in the platform and picking the AI content topic. Additional categories that are already be controlled under the tool cover politics, style, cosmetics, and choreography.
Platform Guidelines
TikTok’s rules mandate creators to identify “realistic” AI-generated content, and it forbids harmful AI impersonations of public figures or critical incidents. Any realistic AI video that is unmarked can be taken down under its terms of service.
Furthermore, the service will from now on add an “AI-made” watermark to content created with its own AI tools, or detected by an cross-platform project for spotting AI-generated material. This step is aimed to stop users from trying to bypass the tagging system.
Educational Initiatives
The company is furthermore establishing a two million dollar educational fund for professionals and institutions to develop informative material about using AI responsibly. This initiative will assist non-profit groups focused on teaching computer science.
Moderation Strategy
The video-sharing app has justified its content review process, amid debate over proposals to cut jobs for a large number of safety staff in Britain. The company is set to cut a significant number of roles in its trust and safety team in the UK capital, with trade unions and internet security professionals voicing alarm that humans are being replaced by machine-led review.
A trust and safety lead clarified that staff-based oversight would continue to be important, but automated tools can protect employees by removing the most graphic content ahead of it is viewed by staff. The company reported a significant reduction in disturbing and violent content handled by content reviewers over the past year due to the use of its machine learning processes.
It is vital to balance people and machines to maintain a secure environment,” the executive stated. “We’ll have humans as involved in the system, but we additionally believe it’s extremely critical to minimize viewing of dangerous material as swiftly, which is the area in which a lot of the machine support is proving valuable work at speed.”
