This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A viral video circulated on social media in Pakistan used AI-generated audio to falsely depict Turkish President Recep Tayyip Erdoğan expressing support for jailed former Prime Minister Imran Khan. The video reused footage from a 2020 visit and overlaid synthetic audio, which was confirmed as likely AI-generated by multiple forensic tools and expert analysis. The disinformation campaign reached hundreds of thousands of viewers across platforms like Facebook and YouTube.
A reported viral video that was circulated in Pakistan featured purported AI-generated audio depicting Turkish President Recep Tayyip Erdoğan expressing support for jailed former Prime Minister Imran Khan during his 2025 visit. The video is described as having reused footage from 2020 and featured AI-generated audio that fabricated remarks Erdoğan never made. The synthetic voice was reportedly confirmed by forensic tools and audio analysis.
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-generated audio fabricated statements by a foreign leader, spreading false information to the public.
The incident involves the creation and viral dissemination of a political deepfake designed to manipulate public opinion and political processes in Pakistan at scale.
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 who intentionally generated and distributed synthetic audio of a political figure to mislead the public after the AI tool was deployed.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
An AI-generated deepfake video falsely depicted Turkish President Erdogan expressing support for jailed Pakistani politician Imran Khan. While the video went viral and garnered hundreds of thousands of views, it represents a minor national security threat that was quickly debunked, highlighting the ongoing but manageable challenge of political deepfakes.
The AI system used to generate the synthetic voice operates as a tool, performing the specific function of voice generation based on direct inputs without autonomous planning or context retention.
Risk Level 3. Limited Risk: The report describes the creation of AI-generated content such as deepfakes, which pose a moderate risk and require transparency obligations under the EU AI Act.
The AI system was used to generate a synthetic replica of Turkish President Erdogan's voice.
The report provides clear, fact-checked details about the synthetic nature of the audio, the specific platforms where it was shared, and the view counts. Multiple forensic tools and expert analyses confirm the AI-generated nature of the clip, leaving little ambiguity about the core events.
None
Technology
AI Voice Cloning Technology
Allowed generation of a highly realistic synthetic voice of President Erdogan.
Acoustic Post-Processing
Artificial reverb was used to blend synthetic audio with authentic background.
Uniform Frequency Generation
Synthetic generator produced constant pitch patterns atypical of human speech.
Human Factors
Malicious Political Motivation
Users intentionally spread the deepfake to influence political narratives.
Lack of Public Verification
Social media users shared the video widely without checking its authenticity.
Process and Methods
Ineffective Platform Moderation
Social media platforms failed to detect and flag the synthetic audio quickly.
Reactive Fact-Checking Process
Fact-checking occurred after the video had already garnered viral reach.
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.
