Meta to Identify AI-Generated Content on its Social Media Platforms

Meta, the parent company of Facebook, Instagram, and Threads, has announced its plans to identify and label AI-generated content across its social media platforms starting from May 2024. This initiative aims at promoting transparency and context while mitigating the risk of suppressing freedom of expression unnecessarily.

The vice president of content at Meta, Monika Bickert, revealed in a blog post that the company plans to commence tagging AI-generated content in May 2024. The «Made with AI» tag, currently used on photorealistic images, will be extended to a wider range of video, audio, and image content. This announcement signifies a noteworthy step towards maintaining authenticity and veracity in the ever-expanding digital realm.

According to Bickert, the AI-generated content will be tagged by the platform if it detects «indicators of AI image» in accordance with industry standards. Also, the content will be marked if users indicate that they are uploading AI-generated content. This new measure is expected to be fully implemented from May 1, 2024.

Meta’s decision to tag AI-generated content comes after consultations with its oversight board. In general, the California-based firm will change how it handles AI-created content. Instead of removing such content as it had been doing, the company will now provide «tags and context». This move is aimed at enhancing transparency and providing more context, which Meta believes is the best way to deal with manipulated content.

While the company is committed to allowing AI-generated content, it maintains a strict stance against any content, human or AI-generated, that violates its rules. Meta warned that any content infringing its policies «against electoral interference, intimidation, harassment, violence… or any other policy in our community standards» will be removed promptly.

To monitor and detect false or misleading AI-generated content, Meta relies on its network of approximately «100 independent fact-checkers». The company had expressed in February its intention to label all AI-generated images as part of its fight against misinformation.

Other tech giants such as Microsoft, Google and OpenAI have also made similar commitments. The main concern is that people could misuse these tools to create political unrest, especially through misinformation or erroneous information, as there are several important elections this year, particularly in the United States.

Beyond these electoral events, the development of generative AI programs has led to a stream of degrading content, according to many experts and regulators. This includes false (deepfake) pornographic images of famous women, a phenomenon that also affects anonymous individuals. The tagging of AI-generated content by Meta is not only a step towards maintaining the integrity of information on its platforms but also an effort to address these concerning issues.

As the AI technology continues to evolve, the need for such measures is expected to increase. Meta’s move to label AI-generated content could set a precedent for other social media platforms, leading to a more transparent and accountable digital environment.