There’s a lot going on at TikTok right now.
As well as online safety updates and new features, the company is introducing sweeping changes to how it moderates the platform’s content.
At the same time, there’s an intense focus on online safety, particularly here in the UK.
With all that going on, Sky News got a rare, exclusive sit-down with one of TikTok’s senior safety executives, Ali Law.
The increasing role of artificial intelligence
One of the biggest changes happening at TikTok is around artificial intelligence.
Like most social media companies, TikTok has used AI to help moderate its platform for years – it is useful for sifting out content that obviously violates policies, and TikTok says it now removes around 85% of violative content without getting a human involved.
Nvidia beats expectations again in defiance of AI bubble fears
Half of novelists fear AI will replace them entirely, survey finds
AI bubble fears take hold of stock markets and bitcoin
Now, it is increasing its use of AI and will be relying less on human moderators. So what’s changed that means TikTok is confident AI can keep young users safe?
“One of the things that has changed is really the sophistication of those models,” said Mr Law, who is TikTok’s director of public policy and government affairs for northern Europe. He explained that AI is now better able to understand context.
“A great example is being able to identify a weapon.”
Whereas previous models may have been able to identify a knife, newer models can tell the difference between a knife being used in a cooking video and a knife in a graphic, violent encounter, according to Mr Law.
“We set a high benchmark when it comes to rolling out new moderation technology.
“In particular, we make sure that we satisfy ourselves that the output of existing moderation processes is either matched or exceeded by anything that we’re doing on a new basis.
“We also make sure the changes are introduced on a gradual basis with human oversight so that if there isn’t a level of delivery in line with what we expect, we can address that.”
Human moderator jobs being cut
That increasing use of AI means TikTok will rely less on its network of tens of thousands of human moderators around the world.
In London alone, the company is proposing to cut more than 400 moderator jobs, although there are reports a number of those jobs will be rehired in other countries.
On 30 October, Paul Nowak, general secretary of the TUC union, said “time and time again” TikTok had “failed to provide a good enough answer” about how the cuts would impact the safety of UK users.
When Sky News asked if Mr Law could ensure UK users’ safety after the cuts, he said the company’s focus is “always on outcomes”.
“Our focus is on making sure the platform is as safe as possible.
“We will make deployments of the most advanced technology in order to achieve that, working with the many thousands of trust and safety professionals that we will have at TikTok around the world on an ongoing basis.”
The UK’s science, technology and innovation committee, led by Labour MP Chi Onwurah, has issued a probe into the cuts, with Ms Onwurah calling them “deeply concerning”.
She said AI “just isn’t reliable or safe enough to take on work like this” and there was a “real risk” to UK users.
However, Mr Law said that, as a parent himself, he is “also highly concerned and highly interested in issues of online safety”.
“That’s why I’m so confident in the changes that we are making at TikTok in terms of content moderation as a whole,” he said.
“The power really comes in the combination of the best technology and human experts working together, and that still is the case at TikTok and it will be going forwards as well.”
Please use Chrome browser for a more accessible video player
New wellness tools
The interview came at the end of an online safety event at TikTok’s Dublin office, its European headquarters.
During the conference, the company announced a number of new features designed to increase user safety, including a new in-app Time and Wellbeing hub for TikTok users.
The hub is designed with the Digital Wellness Lab at Boston Children’s Hospital and gamifies mindfulness techniques like affirmations, not using TikTok during the night and lowering your screentime.
Read more from Sky News:
Meta to block Instagram and Facebook for users under 16 in Australia
Half of novelists fear AI will replace them entirely, survey finds
How violent extremists are thriving online – and why it’s getting harder to catch them
Cori Stott, executive director of the digital wellness lab, said many people use their phones to “set their wellbeing, to reset their emotions, to find these safe spaces, and also to find entertainment”.
The hub was built as part of the TikTok app because young people want wellness tools “where they already are”, without needing to go to a different app, she said.
Still, there are plenty of reports suggesting that phone use and social media has a damaging effect on young people’s mental health… is TikTok trying to solve a problem of its own creation?
Be the first to get Breaking News
Install the Sky News app for free
“If you are a teen on the app, you will load up and find that you have, if you’re under 16, a private profile, no access to direct messaging, a screen time limit set at an hour, [and at] 10pm sleep hour suggestion,” said Mr Law.
“So the experience is one that does try and promote a balanced approach to using the app and make sure that people have the options to set their own guardrails around this,” he said.
“I think the other thing I’d say is that the content on TikTok is, in the main, inspiring, surprising, creative.”
