Russia's Doppelganger Network Used AI-Generated Content to Target Canadian Political Discourse
Russia's Doppelganger network published more than a dozen articles targeting Canadian politics through the "Reliable Recent News" site (2023–2024). OpenAI confirmed the broader operation used ChatGPT for translation and social media comment generation.
Between late 2023 and mid-2024, Russia's Doppelganger disinformation network produced and distributed articles specifically targeting Canadian politics. The operation was run by the Social Design Agency and Structura, operating under the Kremlin's Presidential Administration (Wikipedia, 2024).
A CBC investigation confirmed that a Doppelganger site called "Reliable Recent News" published more than a dozen articles about Canadian politics, in an apparent attempt to undermine support for then-Prime Minister Trudeau and boost his chief rival, Conservative leader Pierre Poilievre (CBC News, 2024). Between November 2023 and August 2024, the broader Doppelganger network operated through more than 700 websites — including fake news sites and clones of legitimate news outlets — across targeted countries (Wikipedia, 2024).
OpenAI confirmed that the Doppelganger operation used ChatGPT to translate articles and generate social media comments for distribution (Wikipedia, 2024). This represents documented use of a commercial generative AI tool in a state-directed disinformation campaign. Whether the articles specifically targeting Canadian politics were AI-generated is not confirmed by available sources.
Global Affairs Canada issued a formal statement on October 28, 2024 on Russian disinformation activities targeting Canada, noting that while Europe and the war in Ukraine appeared to be the primary focus, Canadian social and political issues also featured in the content (Global Affairs Canada, 2024). The House of Commons Standing Committee on Public Safety and National Security (SECU) initiated a study on Russian disinformation campaigns in Canada in September 2024.
Materialized From
Harms
More than a dozen articles targeting Canadian political discourse published through a Doppelganger-operated fake news site, in an apparent attempt to undermine support for the Prime Minister and boost his rival. The broader operation distributed content through more than 700 websites, with OpenAI confirming ChatGPT was used for translation and social media comment generation.
Evidence
3 reports
-
Doppelganger-affiliated 'Reliable Recent News' published 12+ articles targeting Canadian politics
- Global Affairs Canada Statement on Russian Disinformation Primary source
Canadian government formally acknowledged Russian disinformation targeting Canada
-
Overview of Doppelganger operation scope, 700+ fake websites, Social Design Agency / Structura attribution, OpenAI confirmation of ChatGPT use
Record details
Responses & Outcomes
Global Affairs Canada issued a formal public statement acknowledging and condemning Russian disinformation activities targeting Canada, including the Doppelganger network's operations.
The formal statement acknowledged the threat publicly but did not announce new enforcement mechanisms or platform accountability measures.
Editorial Assessment assessed
This is a documented case of a commercial generative AI tool (ChatGPT) being used in state-directed disinformation infrastructure. OpenAI confirmed the use (Wikipedia, 2024). The operation also targeted Canadian politics specifically, with more than a dozen articles about Canadian political figures published through a fake news site (CBC News, 2024). Global Affairs Canada acknowledged the targeting but noted that Canada was not the primary focus of the broader campaign (Global Affairs Canada, 2024).
Entities Involved
AI Systems Involved
ChatGPT was used to translate articles and generate social media comments for the Doppelganger operation, as confirmed by OpenAI
Related Records
- AI-Generated Content and Bot Networks Targeted Canada's 2025 Federal Electionrelated
- PRC Spamouflage Campaigns Used AI-Generated Deepfakes to Target Canadian Politicians and Criticsrelated
Taxonomyassessed
Changelog
| Version | Date | Change |
|---|---|---|
| v1 | Mar 11, 2026 | Initial publication |
| v2 | Mar 12, 2026 | Neutrality review: removed 2 unattributable policy recommendations (OpenAI threat report editorial synthesis, fabricated GAC attribution) per CAIM neutrality policy. |