This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A deepfake video featuring U.S. President Donald Trump expressing support for Biafran independence circulated on Facebook in February 2025. Fact-checkers identified the video as a manipulation of a 2021 BBC clip, generated using the 'ElonTalks' AI platform. The incident highlights the use of synthetic media tools to propagate political disinformation.
In February 2025, a video circulated on Facebook that appeared to show U.S. President Donald Trump expressing support for Biafran independence and calling for the release of Indigenous People of Biafra (IPOB) leader Nnamdi Kanu. However, fact-checkers reportedly traced the visuals to a 2021 BBC video unrelated to Biafra; they noted signs of AI manipulation. The altered clip reportedly used tools such as ElonTalks to generate synthetic audio and video from a user-submitted script.
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 AI system was used to generate highly realistic false statements attributed to a world leader, spreading misinformation on social media.
The AI system was used to generate a political deepfake of a world leader to influence public opinion and support secessionist movements in Nigeria.
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 user intentionally using the deployed ElonTalks AI platform to generate a deceptive deepfake video of Donald Trump for political influence.
Added categories peak at 2: Minor (Epistemic Harm, Indirect)
A synthetic deepfake video of US President Donald Trump falsely expressing support for Biafran independence in Nigeria was generated using the ElonTalks platform and shared on Facebook. While representing an attempt to manipulate political discourse and international relations, the video was quickly debunked by fact-checkers, resulting in minor national security impact.
The AI system ElonTalks operates as a tool, taking a user-provided script and a celebrity video template to deterministically generate a synthetic video without independent decision-making or context retention.
Risk Level 3: Limited Risk. The system is used to generate deepfakes (AI-generated content), which under the EU AI Act are subject to transparency obligations to ensure users are aware the content is synthetic.
The ElonTalks platform is designed to generate realistic synthetic videos and voices of celebrities using templates and user-provided scripts.
The report clearly identifies the AI platform used (ElonTalks), the nature of the manipulation (voice and video synthesis of Donald Trump), and the platform where it circulated (Facebook). Fact-checkers successfully traced the original source video to a 2021 BBC broadcast, leaving no ambiguity about the faked nature of the content.
None. The video is demonstrably a synthetic deepfake.
Technology
Easy AI Video Generation Tools
ElonTalks.com allows realistic video generation from simple text scripts.
Realistic Voice Synthesis
AI synthesis of celebrity voices makes deepfakes highly convincing.
Data Inputs
Use of Real Video Templates
BBC video of Trump from 2021 was used as a template for the deepfake.
Human Factors
Intentional Disinformation
Secessionist group IPOB used the deepfake to spread their political message.
User Credulity
Social media users shared the video without verifying its authenticity.
Process and Methods
Lack of Platform Moderation
Facebook failed to proactively detect and flag the AI-generated fake video.
No Script Verification
ElonTalks.com did not filter or verify the scripts input by users.
Management
Inadequate Safeguards by Platform
ElonTalks.com management failed to implement guardrails against misuse.
Regulatory Environment
Lack of Deepfake Regulations
Absence of laws restricting the creation and sharing of political deepfakes.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
