Network Uses AI Videos to Mimic Creators
McGill researchers identified nearly 300 Instagram and TikTok accounts misusing creators' likenesses in a coordinated influence operation about Canada-US trade tensions.
Updated on Nov. 20, 2026 in Artificial Intelligence

Live Poll
Do you trust the political videos you see online about the Canada-US trade war?
A McGill Centre for Media, Technology and Democracy study published in mid-November identified nearly 300 Instagram and TikTok accounts misusing authentic faces in a coordinated operation. The network published thousands of videos that drew some 20 million views, and TikTok permanently banned the accounts named in the report along with several others.
Why it matters
The operation shows how AI-generated scripts and altered footage of real creators can spread political messaging at scale while appearing to come from genuine personalities. Before TikTok's new deepfake detection tool launched, only four percent of the network's accounts triggered the platform's synthetic content warning.
The network used AI to generate video scripts and digitally altered footage of real creators, then reposted the clips that drew the most engagement. Only four percent of its accounts triggered TikTok's synthetic content warning before the platform's new deepfake detection tool launched.
The players
TikTok
TikTok is a short-form video platform owned by ByteDance that has faced ongoing scrutiny over content moderation and misinformation.
Meta
Meta Platforms owns Instagram and Facebook and has been under pressure from regulators and researchers over coordinated inauthentic behavior on its platforms.
McGill Centre for Media, Technology and Democracy
The McGill Centre for Media, Technology and Democracy is a Montreal-based research group that studies digital media, platforms and democratic processes.
The details
Researchers found thousands of videos featuring personalities ranging from Christian lifestyle influencers to political commentary accounts. Some accounts were registered in China, and posting was concentrated mid-day in Asia, though the study did not establish who was behind the operation.
Timeline
Mid-November 2026: McGill study identifying the network is published.
Friday before November 20, 2026: TikTok launches its new deepfake detection tool.
Following the report: TikTok permanently bans the accounts named, along with several others, for violating rules on misleading content.
The Tech Race
The network's low detection rate exposes the gap between TikTok's newly launched deepfake detection tool and the scale of AI-generated impersonation on short-form video platforms. The takedown and the tool arrived together, signaling that platforms are now competing to catch synthetic content they previously missed.
Users scrolling TikTok and Instagram may encounter familiar-looking creators whose videos were AI-altered without their knowledge. Checking whether a clip carries a synthetic content warning can help viewers judge whether a face and voice are authentic.
The takeaway
Viewers should treat unfamiliar creator videos with some skepticism, especially when they push political messaging. Platform warnings catch only part of synthetic content, so checking who posted a clip and whether the account has a consistent history is a useful habit.
Further reading
For broader context on how synthetic media is shaping online information, explore our Artificial Intelligence coverage.
Source note: This article includes information reported by The Manila times.
Live Poll
Do you trust the political videos you see online about the Canada-US trade war?


