While big U.S. tech players continue dallying in new uses for generative AI, China's Bytedance is getting to work on substantive uses, which could give TikTok an AI usage lead and drive new behaviors through AI tools.

The problem, as most social apps have learned, is that there actually isn't any particularly valuable, right-now use case for generative AI in social apps.

Sure, you can add in a chatbot, but people fundamentally wish to speak with other human beings-that is the justification for the "social" appellation, in any case-and image generation tools seem to have relatively low novelty value, and generative AI post tools also will have a decidedly negative impact on human engagement.

Which brings me to where I think TikTok is operating more sensibly and useful in using features that truly foster engagement.

Although its latest experiment is a bit more dubious about this.

Based on Business Insider, ByteDance has now developed a new AI model that can replicate any person's voice with believable enough accuracy, based on minimal input.

ByteDance's "StreamVoice" system can use just a few utterances to replicate a person's voice in real time, enabling you to replicate virtually any person's speech, here are some examples of what StreamVoice outputs sound like.

Researchers believe it would be used to defraud and because of other fraudulent transactions, among other things, by fraudsters through such applications. To mention but a few, Meta too designs the same, and its "AudioBox" application is now live testing on the web too.
So why invent tools that can mimic people's voices?

According to Meta's AudioBox paper, the tool is set to "lower the barrier of accessibility for audio creation," opening up more opportunity for more people to create audio content.

"Creators could use models like Audiobox to generate soundscapes for videos or podcasts, custom sound effects for games, or any of a number of other use cases,"

Not sure that's all that different to capturing audio directly, but conceptually, it could offer more options to speak and deliver text in your projects, which might facilitate more expansive creativity.

ByteDance is, of course, looking at something similar and, given the popularity of tiktok robot voices already, it could potentially offer more ways to enrich the audio of your clips.

It's just one more step in TikTok's progression toward an evolution of the toolset of AI, including generative AI profile images, improvements in contextual search, and the generation of music in-stream using AI while it's also testing tools like text-to-video and AI chatbots with varying capacities.

And that's not all. In fact, according to a new report published by Bloomberg, TikTok is also now expanding its test of an automated feature that could potentially make every post within the app shoppable, identifying objects in every video and then prompting viewers to "find similar items on TikTok Shop".
That's been in testing for some time, with Insider posting the above image as part of its story on the project last November, while it's also been in testing on Douyin-the Chinese version of the app-since 2019. So ByteDance has had a long time to revise and improve this element ahead of a broader TikTok release.

The incorporation of smarter AI will make it even the more invaluable shopping tool with more products shown to more users and results matched up with preference.
And given the role as the app was already an essential discovery tool for plenty young users, it makes sense for the app to build from this element while encouraging new usage behaviors that are aligned with its broader shopping push.

In essence, TikTok is making the smartest moves of adding AI in a supporting method to help enhance its core use case rather than being tacked on to latch onto the latest tech trend.

As noted, that is one area where other platforms are still catching up, and it will be interesting to see how TikTok will look to stitch in more AI features over time and whether that continues to inform new usage behaviors in the app.