The rise of generative artificial intelligence has created a new challenge for social media platforms: determining what is genuinely produced by humans and what has been created or altered by machines.
In response, platforms such as TikTok and Instagram have introduced labels intended to inform users when content has been generated or significantly modified using AI. The goal is understandable. The problem is that these systems are not always accurate.
An inaccurate AI-generated label can carry consequences far beyond a simple notification. Seven creators have reportedly raised concerns that their human-made work was incorrectly identified as AI-generated.
Register for Tekedia Mini-MBA edition 20 (June 8 – Sept 5, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Such mistakes highlight a growing tension between platforms’ responsibility to provide transparency and their obligation to avoid damaging the reputations of the people who depend on those platforms.
AI detection is inherently difficult. Modern generative tools can produce images, videos, audio and text that increasingly resemble human-created work.
Creators are using sophisticated editing software, cameras, filters, animation techniques and post-production tools that can sometimes produce characteristics associated with synthetic media.
A detection system attempting to distinguish between these categories is therefore operating in an environment filled with uncertainty. Some platforms have acknowledged that their AI detection systems can make mistakes.
That admission is important, but it does not eliminate the problem. For an ordinary viewer, an AI-generated label may appear to be an authoritative statement from the platform. Many users may assume that the creator intentionally used artificial intelligence, even when that conclusion is incorrect.
The reputational impact can be particularly serious for professional creators. Their businesses often depend on trust and authenticity. Photographers, filmmakers, artists, educators and influencers may spend years developing recognizable styles and audiences.
Being associated with undisclosed AI production could make followers question the originality of their work, even when the platform’s classification is wrong.
There is a broader philosophical issue. As AI becomes integrated into everyday creative tools, the boundary between human and machine-generated content is becoming increasingly complicated.
A photograph may be captured by a human but enhanced by an AI-powered editing application. An artist might use generative software for one element while creating everything else manually. A video could contain AI-generated effects alongside hours of human filming and editing.
Instead of treating content as either entirely human or entirely AI-generated, platforms may eventually need more nuanced disclosure systems explaining how AI was used and how confident the platform is in its classification.
Transparency remains essential, particularly as synthetic media becomes more sophisticated and misinformation becomes harder to identify. But transparency should not come at the expense of accuracy.
A misleading label can itself become a form of misinformation when it incorrectly tells millions of users that a creator’s work was produced by AI. The challenge for TikTok, Instagram and other platforms is therefore not merely to detect artificial intelligence.
It is to build systems capable of communicating uncertainty responsibly. Creators deserve tools to challenge incorrect classifications, while audiences deserve meaningful information rather than potentially misleading warnings.
The future of AI transparency will depend on trust. Platforms must acknowledge that detection technology has limits and provide meaningful avenues for correction. Otherwise, an initiative designed to protect users from deception could unintentionally create a different problem.