Sign in
← AI PulseSep 16, 2026

Policy · news · Multi-source brief

OpenAI Disrupts China-Origin Influence Operation "Sponsored Discontent"

OpenAI banned accounts originating from China that used its AI models to generate English social posts and Spanish-language articles, some of which were published by Latin American news sites.

By Illumora Editorial

Source · Sep 16, 2026, 4:02 PM · On Illumora · Sep 16, 2026, 4:13 PM

Media from the primary source — shown here so you can stay on Illumora.

Synthesized from multiple allowlisted primaries on the same event. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →OpenAI Safety — Operation “Sponsored Discontent”: Influence activity
Save

OpenAI announced in its February 2025 report the disruption of a China-origin influence operation it named "Sponsored Discontent." This operation utilized ChatGPT accounts to generate both short comments in English and long-form news articles in Spanish. The activity was consistent with mainland Chinese business hours, and the Spanish-language articles were published by news sites in Latin America.

Key Points

  • OpenAI banned ChatGPT accounts linked to a China-origin influence operation.
  • The operation generated short English-language comments and long-form Spanish-language articles.
  • Activity was consistent with mainland Chinese business hours.
  • Spanish articles were published by news sites in Latin America, particularly in Peru, Mexico, and Ecuador.
  • Some articles were attributed to a Chinese company, Jilin Yousen Culture Communication Co, Ltd.
  • The operation used ChatGPT to translate and expand publicly available Chinese-language articles.
  • Content included criticisms of Cai Xia in English and Spanish articles critical of the US, focusing on social and political divisions.

Context

According to OpenAI, this operation marks the first observed instance of a likely Chinese influence actor successfully publishing articles in mainstream Latin American outlets. The actor generated content by prompting ChatGPT in Chinese. Some of the activity aligned with the "Spamouflage" operation, which OpenAI reported on in May, involving English-language comments and images criticizing Cai Xia. However, the concurrent generation of Spanish articles critical of the US, particularly before the 2024 APEC Forum in Peru, indicated a distinct effort. OpenAI also noted that some publications might have resulted from financial arrangements, with one identified article listed as "sponsored."

Why It Matters

This case highlights how AI models can be leveraged by state-linked actors to produce and disseminate influence content across multiple languages and platforms, potentially reaching wider audiences through mainstream media. Builders and practitioners should note the evolving tactics of influence operations, which now include using AI for content generation and translation to penetrate established news channels, posing new challenges for content moderation and platform security.

What To Do

  • Note the February 2025 report from OpenAI Safety detailing the "Sponsored Discontent" operation.
  • Watch for similar patterns of AI-generated content appearing in mainstream media outlets.
  • Review OpenAI's other reports on influence operations, such as "A2Z" (October 2024), "Uncle Spam" (June 2025), and "Stop News" (October 2024), to understand diverse tactics.
  • Consider how AI models are used for translation and content expansion in influence campaigns.