Stay up to this point with free updates
Just register Artificial intelligence myFT Digest – delivered straight to your inbox.
TikTok will turn out to be the primary social media platform to robotically label some artificial intelligence-generated content, as rapid advances in generative AI heighten concerns in regards to the spread of online disinformation and deepfakes.
Online groups similar to Facebook operators Meta and TikTok already require users to reveal whether realistic images, audio or video content is created using AI software.
Chinese company ByteDance's viral video app went a step further on Thursday, announcing the launch of its own features to make sure that videos it could possibly discover as AI-generated are labeled as such. This includes content created using Adobe's Firefly tool, TikTok's AI image generators, and OpenAI's Dall-E.
“The challenge is that we all know from most of the experts we work with that there may be a rise. . . harmful AI-generated content,” said Adam Presser, head of operations and trust and safety at TikTok.
“This is admittedly vital for our community because authenticity is admittedly one in all the weather that has made TikTok such a vibrant and joyful community.” . They need to give you the option to grasp what was created by a human and what was improved or generated with AI.”
Social media platforms similar to TikTok, Meta, However, the platforms have come under fire for allowing users to be flooded with low-quality AI-generated spam content.
In a yr of major elections around the globe, these firms are also under pressure to implement protections against misleading deepfakes, curb covert influence operations and ensure they properly moderate content while remaining impartial.
Earlier this week, TikTok and its parent company ByteDance filed a lawsuit against the U.S. government, difficult a law intended to force a sale or ban on the app. Lawmakers had expressed concerns that the platform could promote disinformation and propaganda.
On Thursday, TikTok said it will join a coalition of technology and media groups led by Adobe which might be integrating so-called content credentials into AI-generated products.
This technology embeds a digital fingerprint into multimedia AI content, together with other identifying information similar to when, where and who created the fabric. TikTok uses these indicators to robotically flag when content is created using AI products.
OpenAI announced Tuesday that it will join the coalition and embed fingerprint technology into all images generated by its Dall-E 3 imaging model. Finally, the ChatGPT maker said it would also embed this into its Sora video generation model when it’s widely released.
Major technology firms, including Google, Microsoft and Sony, are exploring embedding the technology into their AI tools.
Meta announced earlier this month that it will begin applying the “Made by AI” label to AI-generated content by detecting invisible markers inserted by groups similar to Google, OpenAI, Microsoft, Adobe, Midjourney and Shutterstock turn out to be. The Facebook owner also stated that it’s developing classifiers for deepfake detection.
Experts have argued that malicious actors or sophisticated disinformation groups would likely use open-source AI generation tools to create deepfakes, making them harder to detect and avoid being flagged by digital fingerprints and watermarks.
Tech firms argue that their efforts represent step one toward solving the issue.
Dana Rao, general counsel and chief trust officer at Adobe, said: “The premise of this solution is: If you must be transparent and have authentic, transparent conversations together with your public, this tool lets you try this in a world where all the things “It’s possible that digital things will be manipulated.”
