Chinese state-affiliated actors alleged to run AI-driven diplomatic personas on social media

In 2023 researchers reported coordinated influence campaigns that used AI-generated avatars and synthetic personas to promote pro-China narratives across YouTube and other social platforms. Independent research groups (ASPI, Graphika) documented instances of AI-enabled video and avatar use and flagged the campaigns as coordinated and inauthentic; press outlets subsequently reported the findings. Attribution to Chinese state-affiliated actors is reported by researchers and framed as alleged.

Chinese state-affiliated actors (alleged) · Incident Dec 1, 2023 · Indexed Jun 10, 2026 · 3 sources

Automated AI-generated avatars and synthetic personas were used to seed and amplify pro-China narratives across platforms.
What
In 2023 researchers reported coordinated influence campaigns that used AI-generated avatars and synthetic personas to promote pro-China narratives across YouTube and other social platforms.
Incident date
Dec 1, 2023
Who
Chinese state-affiliated actors (alleged)
Failure mode
Tool Misuse
AI surface
Agentic Workflow
Severity
High

What happened

Researchers documented coordinated inauthentic influence campaigns during 2023 that promoted pro-China and anti-US narratives on YouTube and other platforms, using AI-generated video, avatars, and synthetic personas to publish and amplify content. ASPI reported a coordinated campaign active on YouTube, and Graphika documented instances where video footage and AI-generated avatars were used by pro-Chinese operations. Press outlets covered the research findings and described instances of AI avatars being used to present videos and commentary.

What broke inside the model

Failure path · mode profile · Tool Misuse
  1. 01 · TriggerThe agent selects the correct tool.
  2. 02 · Model stepIt fills the call with the wrong arguments.
  3. 03 · Control gapNo validation checks the arguments first.
  4. 04 · FailureThe tool runs against the wrong target.
  5. 05 · ConsequenceThe wrong record, account, or system is hit.

At the tool call, the arguments point at the wrong target.

Generative AI tools were used to create convincing synthetic personas and avatars that platforms failed to detect or remove promptly, enabling coordinated amplification of targeted narratives. The failure combined misuse of readily available synthetic-media tools with platform detection and authenticity controls that were not effective at identifying coordinated, AI-driven inauthentic behavior.

Public visibilityMedium
Regulatory exposurePossible
Customer impactMany customers
Financial impactUnknown
Time to disclosureMonths
  1. PrimaryShadow Playaspi.org.au
  2. PrimaryDeepfake It Till You Make It (Graphika report)public-assets.graphika.com
  3. PressPro-China campaign targeted YouTube with AI avatarstheregister.com
Permalinkhttps://failureindex.ai/failures/chinese-state-affiliated-actors-alleged-run
CitationAI Failure Index. "Chinese state-affiliated actors alleged to run AI-driven diplomatic personas on social media" (FI-0443). Realm Labs. https://failureindex.ai/failures/chinese-state-affiliated-actors-alleged-run (indexed Jun 10, 2026).
Share cardA branded image of this record for posts and slides.

Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0443. Full dataset at /data.

Note from Realm Labs, the Index steward

How Realm would have caught this

Controls for this failure mode
  • OmniGuard
  • AgentRealm

Realm can inspect a tool call against the user's actual intent before it runs, and hold calls whose arguments or target do not match what was asked, so the wrong tool or the wrong arguments never reach the system of record.