Purported Deepfake Images of Gráinne Seoige Reportedly Circulated During Ireland's 2024 General Election Campaign

A reported deepfake abuse campaign targeted Gráinne Seoige during her run in Ireland's 2024 general election campaign, when pornographic, purportedly AI-generated images of her were reportedly circulated online. Seoige later described the experience as humiliating and traumatic, and Gardaí said they were investigating the distribution of the material.

Risk classification

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

The incident involved the creation and circulation of non-consensual pornographic deepfakes of a specific victim to target, humiliate, and politically disadvantage her.

Additional risk subdomains

• 1.2 Exposure to toxic content: The incident involved the creation and online spread of sexually explicit, non-consensual synthetic imagery.
• 4.1 Disinformation, surveillance, and influence at scale: The deepfakes were distributed as part of a coordinated political campaign to influence public opinion and sabotage an election.

Causal factors

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

The risk and harm were caused by the intentional decision of human political opponents to generate and distribute deepfake pornography to damage the victim's campaign.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The incident involves AI-generated content, specifically deepfakes, which are subject to transparency obligations under the EU AI Act to ensure users are aware the content is synthetic.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Image Generation; Image Generation
• Behaviour type: Tool
• Alleged developer: Unknown image generator technology, Unknown deepfake technology developers
• Alleged deployer: Unknown disinformation actors, Unknown deepfake creators
• Alleged harmed parties: Voters in Ireland, Gráinne Seoige, General public of Ireland, General public, Epistemic integrity, Electoral integrity

Harm severity

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

Financial

Reported: Yes, the report describes unquantified financial losses due to loss of employment.

Directly caused: N/A

Indirectly caused: The victim reported being left effectively unemployed with no TV or MC bookings for the year due to the fallout of the campaign.

Inferred additional harm: Significant unquantified loss of income from lost media and hosting contracts due to reputational damage.

Malicious content

Reported: Yes, the report describes the creation and spread of toxic synthetic pornographic images.

Directly caused: AI-generated pornographic images of the victim were created and circulated online.

Indirectly caused: N/A

Inferred additional harm: The explicit images were likely viewed by thousands of social media users and voters in her constituency.

Differential treatment

Reported: Yes, the victim highlighted that this type of abuse disproportionately targets women and children.

Directly caused: N/A

Indirectly caused: The victim was subjected to gender-based online harassment and sexualization that male political candidates typically do not experience.

Inferred additional harm: N/A

Democracy

Reported: Yes, the report describes an attempt to sabotage a democratic election campaign.

Directly caused: N/A

Indirectly caused: The deliberate circulation of deepfake pornography by political opponents interfered with a democratic election by damaging a candidate's campaign and discouraging her from media participation.

Inferred additional harm: N/A

Psychological

Reported: Yes, the report explicitly describes severe psychological harm to the victim.

Directly caused: Grainne Seoige experienced deep trauma, shame, and humiliation for almost a year and a half, describing it as the most shocking thing that has ever happened to her.

Indirectly caused: N/A

Inferred additional harm: N/A

Epistemic

Reported: Yes, the report describes the creation of false synthetic images.

Directly caused: The generation of fabricated visual evidence falsely depicting the victim in explicit scenarios.

Indirectly caused: N/A

Inferred additional harm: N/A

People affected

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

Potential causes

Management

• Safety Investment Deficit: Platforms prioritize growth and engagement over robust safety systems.

Technology

• AI Deepfake Image Generators: Easy access to AI tools allows creation of realistic non-consensual images.
• Algorithmic Amplification: Social media algorithms spread sensational deepfake content rapidly.

Data Inputs

• Unprotected Public Images: Public photos from social media were harvested to generate deepfakes.

Human Factors

• Malicious Political Intent: Opponents deliberately used deepfakes to sabotage a political campaign.
• Victim Silence and Shame: Trauma and social shame delayed public reporting of the digital abuse.

Process and Methods

• Slow Content Moderation: Platforms failed to quickly detect and remove harmful deepfakes.
• Inadequate Reporting Channels: No effective real-time mechanism to halt the spread of flagged images.

Regulatory Environment

• Untested Deepfake Laws: Legislation like Coco's Law is untested for adult manipulated deepfakes.
• Lack of Platform Liability: No legal duties or penalties exist for platforms failing to act on deepfakes.
• Limited Enforcement Powers: Police lack real-time powers to quickly stop the spread of digital abuse.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide clear, consistent, and detailed accounts of the incident, including direct quotes from the victim, details of the ongoing police investigation, and the specific impacts on her career and mental health. There is no ambiguity about the nature of the AI's role as a generator of deepfake pornography.
• Ambiguities identified: The specific AI tool or generator used to create the images is not identified.

During Ireland's 2024 general election, candidate Grainne Seoige was targeted by political opponents using AI-generated non-consensual sexual deepfakes to sabotage her campaign. The incident represents a minor national security impact, highlighting the evolving threat of localized, AI-enabled information manipulation and election interference, which was managed under standard domestic law enforcement procedures.

• Overall national security impact: Minor
• Response level: Moderate
• Scope: Single nation
• Primary target: Ireland
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The immediate election event has passed, but the incident represents an ongoing strategic concern regarding the use of deepfakes in democratic processes.
• Autonomy: Human-controlled. The AI system was used purely as a tool by human operators to generate specific images, with no autonomous decision-making or execution.
• Novelty: Evolved capability. While deepfakes are an established technology, using non-consensual sexual deepfakes specifically to target and sabotage a political candidate in an election represents an evolved and highly targeted threat vector.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or kinetic assets were targeted or affected in this incident.
• Information security: Minor. Involves AI-generated deepfake disinformation targeting a political candidate, representing localized information manipulation rather than a large-scale or state-sponsored campaign.
• Sovereignty: Minor. The incident represents an attempt to interfere with democratic processes by targeting a candidate in a national election, but the scale was localized and managed under standard law enforcement procedures.
• Economic security: Negligible. No strategic technology, critical supply chains, or financial systems were impacted; financial loss was limited to the individual level.
• Societal stability: Minor. The incident involved gender-based harassment and violation of personal dignity through non-consensual synthetic imagery, highlighting societal risks of AI but on an individual scale.