Hong Kong to face hurdles in enforcing rules on labelling AI-generated content: experts

Hong Kong could struggle with enforcing regulations on labelling artificial intelligence-generated content on social media platforms, experts have said, after mainland China rolled out a new law this week to combat misinformation, deepfakes and copyright issues.

Experts told the Post that Hong Kong’s market was too small to adopt an independent set of laws to regulate the use of generative AI, but a rising number of deepfake scams and improper use of the technology reflected a growing urgency for tighter regulation.

On Monday, China’s top internet watchdog, the Cyberspace Administration of China, along with three other government departments, rolled out a new law that mandated the labelling of all artificial intelligence-generated content online.

It stated that explicit markings must be clearly visible to users, while implicit identifiers – such as digital watermarks – should be embedded in the metadata.

Major Chinese social media platforms, including Tencent, Douyin, Kuaishou, Bilibili and DeepSeek, have launched new features to abide by the law issued in March.

Some platforms also penalise users who fail to identify AI-generated content through measures such as limiting traffic, removing content, banning accounts and cancelling revenue.