WhatsApp is preparing a new option that will let channel administrators mark certain posts as containing AI-generated media. The feature appears in the latest Android beta and is framed as a response to transparency rules taking effect in the European Union. Once an administrator shares an update that includes images, videos or other media created or altered with AI tools, they can press and hold the message to open a context menu and select “Add AI content label.” The label then appears on the message bubble so followers can see it at a glance. After the label is applied it may not be removable.
The process mirrors the existing flow for flagging sponsored content, suggesting WhatsApp is reusing familiar interface patterns rather than inventing an entirely new workflow. Current descriptions indicate the requirement focuses on visual and audiovisual media; AI-generated text does not appear to fall under the same obligation. Whether the tool will roll out globally or remain limited to jurisdictions that mandate such disclosures is still unclear. An iOS version is expected to follow given the regulatory driver.
Platforms have faced mounting pressure to make synthetic media more identifiable as generative tools become widespread. The European rules push providers toward clearer labeling so users can distinguish human-created material from machine-generated content. Other large services have experimented with similar disclosures, yet adoption and consistency remain uneven. Relying on channel administrators to apply the labels introduces an obvious limitation: the system depends on voluntary or semi-voluntary compliance rather than automatic detection. An administrator who overlooks or deliberately omits the label leaves followers without the intended signal.
Channels themselves have grown into a significant distribution surface on WhatsApp, used by news outlets, public figures and interest groups to reach large audiences outside traditional social feeds. Adding an AI marker gives those audiences one more piece of context, but it does not solve deeper questions about verification, provenance or the rapid evolution of generative tools that can produce increasingly realistic output. Enforcement will likely vary by country, and the technical details of what counts as “created or edited using AI tools” may prove harder to police than the interface suggests.
For now the feature remains in testing. Its eventual reach and practical impact will depend on how consistently administrators use it and how clearly the labels are understood by ordinary users. In markets where the rules are binding the option may become routine; elsewhere it could stay optional or appear only when local law demands it. The change reflects a broader industry shift toward visible disclosure of synthetic media, even as the underlying technology continues to outpace the labeling systems built around it.
