Opinion

From Content Moderation to Crisis Intervention: The New Challenge for TikTok and Big Tech

By Manika

Posted on August 5, 2026

8 min read

The original concept of social media was to bring people together, to entertain, to inspire digital creativity. However, with time they have evolved into something much more sophisticated. Apps like TikTok are supposed to come across and acknowledge crises in real-time, in situations that match more closely to those of public safety than standard content moderation.

But as technology becomes increasingly capable, now the question is raised if it should be technology removing harmful content, or be the tech company taking a bigger role when lives can be at risk.

Limits of Traditional Content Moderation

Content moderation has been about spotting plagiarism, hate speech, and nude images as well as identifying spam back then, and it still is to some extent. Platforms used artificial intelligence to process millions of posts a day, with human moderators taking on ironclad review of difficult posts, which were beyond the scope of AI.

Things have been tilted and the equation has been changed with Livestreaming. Whilst showing recorded videos, the live shows do not allow for any pre-screening. Every minute matters when someone has a crisis or is live broadcasting to others about harming themselves and talking during a mental health crisis. Thousands of people online can see a disturbing incident without intervention before it finally gets moderated.

Why Livestreams Present a Different Challenge?

The example demonstrates the increasing challenges for real-time content moderation on TikTok Live. While automated systems are able to detect specific keywords or visual cues, they are less likely to pick up on context, sarcasm, slang or abrupt shifts in behavior.

The impact is not only on the person who is broadcasting it, it will be on other people whose lives are affected. Live streams of any crises can start to cause panic among streamers and fans, can encourage toxic online interactions, and can be quickly picked up on more platforms than one by screencaps and re-posts. The act of removing the original livestream may be a first step toward a response.

AI Alone Cannot Handle Human Emergencies

AI has revolutionised the velocity of content moderation, although there are its limitations in the face of complicated human scenarios. Interpretation, empathy, and a good judgment are often needed during a live stream that isn’t consistently provided by current AI systems, and might involve emotional distress.

Protests have grown loud over the past few years about the greater dependence on automation on platforms as companies have cut down on Trust & Safety teams. AI is very good at spotting policy breaches, but in crisis situations trained human reviewers are often required to make a determination about the crisis and upgrade exceptions accordingly.

Crisis Intervention Is Becoming Part of Platform Responsibility

Social mediapersons’ role is not only to remove harmful content but is slowly expanding. As more platforms realize the overlap and need to communicate with emergency services, law enforcement and mental health agencies when credible threats to life arise while broadcasting, they are beginning to strike that deal.

The shift is part of a trend that digital platforms be seen as responsible actors in society. Instead of requiring enforcement of community guidelines, companies now need to be in response mode, taking measures to maintain data and make it possible for timely intervention when needed.

Regulators Are Raising the Stakes

Through legislation, governments are affirming these expectations. UK’s regulation rules like the Online Safety Act and the European’s Digital Services Act, are stakeholders that require more duty of care particularly for vulnerable content and users.

The laws highlight key issues such as speedy response mechanisms, transparency, parental safety tools and effective risk management. Therefore, the response to crises is no longer seen just as a societal duty; but also as a legal one.

Future Requires Human-AI Collaboration

It’s not about whether AI or humans should be used for social media moderation; it’s about how both can co-exist and thrive together. AI systems can identify unusual operations and set priorities on potential crises, while a team of trained humans can make use of the expertise to recognize a genuine crisis from a fake one.

This hybrid will be instrumental to making digital platforms safer. Whilst warning signs can be detected using technology, relevant action will require a coordinated, joint institutional response with clear emergency protocols, relying on human expertise.