TikTok Boosts AI Content Detection and Literacy to Combat Spam

TikTok has outlined new initiatives aimed at identifying and mitigating the impact of AI generated content, particularly spam. These efforts include expanded literacy programmes, enhanced detection systems, and deeper transparency partnerships within the technology sector. The company seeks to strengthen safeguards against misuse.

The platform is expanding its AI literacy efforts through new educational resources. It has collaborated with industry partners NAMLE and Henry Ajder to produce a guide for responsible AI tool usage. An in app hub will also launch, offering practical skills to users on how to identify AI generated content when searching for related terms.

The platform acknowledges that while AI offers storytelling opportunities, it can also be exploited to mass produce spam, potentially overshadowing authentic content creators. TikTok has historically prohibited spam and employed technology to remove it at scale, removing more than 86 million fake accounts in the initial three months of the current year. The evolving nature of such content necessitates continuous updates to protective measures.

Tiktok has plans to improve its detection systems, specifically target accounts dedicated to posting AI generated spam on subjects that could compromise public trust or wellbeing. This includes content related to politics and current events, financial advice, and medical information. But, as creator engagement with AI increases, TikTok continues to enhance its labelling technologies and related partnerships.

The platform has also adopted C2PA Content Credentials, an AI transparency technology designed to indicate when content has been generated or significantly edited by artificial intelligence and has applied labels to over 3 billion videos identified as AIGC. This process uses a combination of Content Credentials, creator labelling tools, and proprietary invisible watermarking technology.

The company will also continue to test features that allow users more control over the AIGC they encounter. Manage Topics is one such feature, enabling individuals to choose the quantity of AI generated content they view.

• TikTok is expanding AI literacy initiatives and educational resources.

• Improved detection systems are being tested to combat AI generated spam.

• The platform has joined the C2PA Steering Committee to promote transparency technology.

• Over 3 billion videos have been labelled as AI generated content.

Source: TikTok Newsroom