This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A deepfake video depicting a fabricated physical altercation between Bougainville President Ishmael Toroama and Papua New Guinea Prime Minister James Marape circulated online. The video caused public confusion and was condemned by President Toroama as a dangerous attempt to destabilize diplomatic relations and peace efforts. The President called for regulatory action against the misuse of AI for misinformation.
A purported AI‑generated deepfake video circulated online falsely depicting a physical altercation between Bougainville President Ishmael Toroama and Papua New Guinea Prime Minister James Marape. The footage reportedly caused public confusion and risked damaging relations between Bougainville and Papua New Guinea. Toroama reportedly condemned the video as false and malicious, as well as potentially destabilizing, urging stronger action against misleading AI‑generated content.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim of manipulating political processes, public opinion, and behavior.
Also classified as
The deepfake video created highly realistic but entirely fabricated political events, leading to widespread public confusion and false beliefs.
The incident involved the intentional creation and wide social media distribution of a deepfake video depicting a fake fight between two political leaders, aiming to influence public opinion and destabilize diplomatic relations.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was caused by a human actor intentionally using a deployed AI deepfake generation tool to create and distribute a misleading video of political leaders.
Added categories peak at 2: Minor (Epistemic Harm, Direct)
An AI-generated deepfake depicting a physical fight between the leaders of Papua New Guinea and Bougainville circulated online. The incident represents a substantial information warfare and sovereignty threat, aiming to destabilize sensitive diplomatic relations and ongoing peace negotiations in a highly volatile political context.
The report describes the video as 'AI-generated' and 'digitally manipulated', indicating that an AI system was used as a tool by a human creator to produce the specific fabricated content.
Limited Risk: The system is described as generating 'AI-generated content such as deepfakes', which falls under Risk Level 3 due to transparency obligations to ensure users are aware the content is AI-generated.
The AI system was used to create a digitally manipulated, fabricated video of political leaders (Deepfake Video Generation) which was then shared widely on social media platforms (Social Media Content Generation).
The report clearly describes the nature of the incident (a political deepfake video), the reactions of the affected leaders, and the societal impact. While the specific AI tool used is not named, the classification of the risk domain and harms is highly certain based on the detailed descriptions of the video's content and its political consequences.
None. The incident is unambiguously a political deepfake designed to mislead.
Technology
AI Deepfake Generation Tools
Advanced AI tools allowed creation of realistic fabricated video of leaders.
Social Media Distribution Algos
Platforms allowed rapid, unchecked sharing of misleading AI content.
Human Factors
Malicious Content Creation
Content creators deliberately misused AI to spread false information.
Public Confusion and Sharing
Social media users shared the video without verifying its authenticity.
Process and Methods
Inadequate Platform Moderation
Social media platforms lacked robust steps to curb fake AI content.
Management
Platform Accountability Deficit
Social media management failed to prioritize ethical AI safety controls.
Regulatory Environment
Lack of Regulatory Oversight
Regulatory bodies like NICTA lacked proactive measures to prevent spread.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
