Northern Ireland MLA Cara Hunter Allegedly Targeted by Deepfake Pornography Ahead of May 2022 Assembly Election

Sometime in April 2022, reportedly some weeks before the May 5, 2022 Northern Ireland Assembly election, Member of the Legislative Assembly Cara Hunter was targeted in a deepfake pornography incident. A fake video depicting her in explicit scenarios is reported to have circulated widely on WhatsApp. Law enforcement have as of yet been unable to trace the origins of the video, according to reports. Additionally, it is alleged that about six months after the initial incident, she was the target of at least 15 more AI-generated deepfakes depicting her likeness in underwear.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involved the creation and viral distribution of non-consensual sexual deepfakes to target, humiliate, and politically manipulate the reputation of a specific victim.

Additional risk subdomains

• 1.2 Exposure to toxic content: The incident exposed the victim, her family, and the public to highly explicit, non-consensual pornographic synthetic media.
• 3.1 False or misleading information: The deepfake video generated and spread false information about the victim's actions, leading to inaccurate beliefs among voters.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The incident was initiated by a malicious human actor who intentionally used a deployed AI deepfake tool to generate and distribute explicit content targeting the victim.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The EU AI Act classifies AI-generated content such as deepfakes under Limited Risk, which imposes specific transparency obligations such as notifying users that the content is AI-generated.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Deepfake Video Generation; Deepfake Image Generation
• Behaviour type: Tool
• Alleged developer: Unknown deepfake technology developers
• Alleged deployer: Unknown deepfake creators
• Alleged harmed parties: Voters in Northern Ireland, Truth, Epistemic integrity, Electoral integrity, East Londonderry constituents, Democracy, Cara Hunter

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Substantial, indirect Substantial
• Differential treatment: direct Minor, indirect Minor
• Civil rights: direct Minor, indirect Minor
• Democracy: direct Minor, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Minor, indirect Minor
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes

Directly caused: An explicit 40-second pornographic deepfake video was created and shared thousands of times across WhatsApp and other social media platforms.

Indirectly caused: The viral spread of the video prompted a wave of unsolicited, sexually explicit, and abusive messages from men on social media.

Inferred additional harm: The explicit content likely remains hosted on various online adult forums and deepfake websites, continuing to expose users to the toxic material.

Differential treatment

Reported: Yes

Directly caused: The victim was targeted with highly sexualized deepfakes specifically because she is a young female politician, reflecting a broader pattern where 99 percent of deepfake targets are women.

Indirectly caused: She faced hostile public reactions, mocking, and street harassment that male political candidates do not typically experience.

Inferred additional harm: The threat of such targeted sexual harassment creates systemic barriers that deter women from participating in public life and political campaigns.

Civil rights

Reported: Yes

Directly caused: The non-consensual creation of explicit sexual imagery using her likeness violated her fundamental right to dignity, privacy, and bodily autonomy.

Indirectly caused: Her ability to campaign safely and express her political voice was severely compromised by the targeted harassment campaign.

Inferred additional harm: Widespread synthetic sexual abuse represents a systemic threat to the civil rights and safety of women in digital and physical public spaces.

Democracy

Reported: Yes

Directly caused: The deepfake campaign was used as a digital weapon to derail the victim's electoral ambitions and sabotage her political career just weeks before an election.

Indirectly caused: The campaign forced her to alter her public interactions and nearly altered the election outcome, which she won by only 14 votes.

Inferred additional harm: The proliferation of political deepfakes threatens the integrity of democratic elections by manipulating public opinion through fabricated evidence.

Psychological

Reported: Yes

Directly caused: Cara Hunter experienced severe stress, anxiety, and trauma, describing it as the most horrific and stressful time of her life, which still affected her mental health three years later.

Indirectly caused: Her family members, including her father and uncle, experienced significant distress, confusion, and worry upon seeing the convincing fake video.

Inferred additional harm: Other female politicians and victims of similar deepfake campaigns in the region likely suffered comparable severe psychological distress and anxiety.

Epistemic

Reported: Yes

Directly caused: The creation of a highly convincing fake video led many members of the public and even family members to believe she had actually participated in a pornographic film.

Indirectly caused: The false rumors persisted for years, forcing the victim to repeatedly defend her reputation and explain that the video was entirely fabricated.

Inferred additional harm: The widespread availability of realistic deepfakes erodes public trust in visual media, undermining the shared consensus reality necessary for informed decision-making.

People affected

• Occurrences reported: 1
• People reportedly harmed: 2
• People reportedly exposed: 2

Potential causes

Management

• Inadequate cyber-crime funding: Police departments lack the technology and expertise to trace deepfakes.
• Prioritizing growth over safety: Social media giants fail to implement proactive deepfake protections.
• Lack of candidate preparation: Electoral bodies do not prepare candidates for digital smear campaigns.

Technology

• Rapid AI face-swap advancement: AI makes creating highly realistic face-swaps fast and accessible to anyone.
• Accessible nudification apps: Cheap or free apps allow users to virtually undress any photo with ease.
• Lack of AI watermarking: Absence of mandatory markers on synthetic media hinders verification.

Data Inputs

• Stolen social media photos: Publicly available images are harvested to train deepfake models.
• Innocent source videos: Ordinary internet videos are used to map faces onto explicit acts.

Human Factors

• Targeted political misogyny: Deepfakes are weaponized to humiliate and silence female politicians.
• Public apathy to deepfakes: Some people view non-consensual deepfakes as harmless entertainment.
• Victim shaming and isolation: Social stigma leads to victims being shunned in their communities.

Process and Methods

• Slow platform moderation: Social media networks fail to quickly remove explicit synthetic content.
• Encrypted viral distribution: WhatsApp encryption allows rapid, anonymous, and untraceable sharing.

Regulatory Environment

• Legislative gaps on creation: Existing laws target distribution but fail to ban deepfake creation.
• Slow legislative processes: Government regulations cannot keep pace with rapid AI advancements.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent, detailed first-hand accounts from the primary victim, Cara Hunter, regarding the timing, dissemination, and real-world impacts of the deepfake video. The context of the 2022 Northern Ireland Assembly election and the subsequent legislative efforts are well-documented across multiple reputable news sources.
• Ambiguities identified: The exact AI tool or app used to generate the deepfake is not identified, and there was initial uncertainty by the victim on whether it was a deepfake or a lookalike actress, though later confirmed as AI-generated.
• Alternative interpretations: None. The reports consistently identify the video as a non-consensual deepfake used for targeted political and sexual harassment.

A local Northern Irish politician was targeted with a non-consensual deepfake pornography campaign weeks before an election, leading to harassment and reputational damage. The incident highlights the growing threat of AI-enabled political sabotage and gender-based harassment, though its impact remained localized and did not escalate to a systemic national security crisis.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United Kingdom
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The immediate electoral crisis has passed, representing an ongoing strategic concern regarding the regulation and policing of synthetic media.
• Autonomy: Human-controlled. The AI was used strictly as a tool by a human operator to generate the deepfake content, with no autonomous decision-making by the system.
• Novelty: Evolved capability. While deepfakes existed, using highly convincing AI-generated sexual content to target and sabotage a political candidate in an active election represents an evolved threat.

Impact by dimension

• Physical security: Negligible. No threat to physical systems, critical infrastructure, or kinetic military capabilities was identified in this incident.
• Information security: Minor. Targeted deepfake campaign used for localized political sabotage, but lacked the scale, state-actor backing, or national intelligence impact to threaten broader national security.
• Sovereignty: Minor. Localized attempt to disrupt a regional democratic election by targeting a candidate, though it did not cause systemic disruption to national government functions.
• Economic security: Negligible. No threat to financial systems, strategic technologies, or economic stability.
• Societal stability: Minor. Led to localized harassment and violated the victim's civil rights, reflecting broader societal concerns about synthetic media targeting women, but did not cause widespread civil unrest.