Do not involve my daughter in political matters
En Colombia, donde las campañas presidenciales de 2026 ya muestran sus tensiones más profundas, la inteligencia artificial fue utilizada para fabricar una alianza inexistente entre dos candidatas y para involucrar a una niña en una disputa política que no le pertenece. Las senadoras Paloma Valencia y Claudia López, junto con la esposa de esta última, Angélica Lozano, rechazaron las imágenes y afirmaciones falsas difundidas por una usuaria de X afín al candidato Abelardo de la Espriella. El episodio revela algo más que una táctica sucia de campaña: señala el momento en que la tecnología deja de falsificar solo hechos y comienza a fabricar intimidades, convirtiendo la cercanía cotidiana en evidencia de conspiraciones que nunca existieron.
- Una usuaria de X publicó imágenes generadas por IA que mostraban a Valencia y López como aliadas secretas, añadiendo la falsa afirmación de que Lozano era madrina de la hija de Valencia, antes de borrar la publicación.
- Valencia respondió con urgencia y tono personal, trazando una línea inequívoca: su hija no debe ser arrastrada a ninguna estrategia de ataque político, sin importar el medio utilizado.
- López denunció el uso de desinformación fabricada con IA como herramienta de campaña y calificó la táctica de desesperada y burda, mientras negaba tanto la alianza como el vínculo familiar inventado.
- Lozano desmontó la acusación con lógica directa: haber visto a la hija de una colega en el Congreso no crea ningún lazo de parentesco, y así lo ocurre con los hijos de cualquier otro legislador.
- El incidente expone una frontera que las tres mujeres exigen respetar: los menores de edad no son instrumentos electorales, y la IA no puede convertirse en el mecanismo para manufacturar vínculos falsos entre adversarios políticos.
Una usuaria de X que apoyaba al candidato presidencial Abelardo de la Espriella publicó imágenes generadas por inteligencia artificial que presentaban a Paloma Valencia y Claudia López como aliadas políticas ocultas. Junto a las imágenes, afirmó que Angélica Lozano, esposa de López, era madrina de Amapola Rodríguez, la hija de Valencia, sugiriendo que ambas candidatas habían engañado al público. La publicación fue eliminada, pero no antes de que las aludidas la vieran.
Valencia reaccionó de forma directa y personal: exigió a de la Espriella y sus seguidores que dejaran a su hija fuera de la contienda política. Su mensaje no fue una declaración institucional sino la advertencia de una madre que traza un límite.
López adoptó un enfoque más amplio. Calificó la publicación como una pieza de desinformación fabricada con IA, negó tanto la supuesta alianza como el vínculo familiar inventado, y criticó la campaña de de la Espriella como desesperada. También pidió que se dejara de involucrar a menores en estrategias electorales.
Lozano intervino para desmontar la acusación con precisión: había visto a la hija de Valencia en el Congreso, como también había visto a los hijos de otros legisladores de distintos partidos. Con una pregunta retórica, señaló el absurdo de la lógica acusatoria: ¿acaso eso la convertía en madrina de todos ellos?
Cuando el columnista Sixto González le pidió a Valencia que aclarara públicamente si Lozano era madrina de su hija, la respuesta fue rotunda: no. Valencia le pidió que difundiera esa negativa. Lozano volvió a pronunciarse, esta vez para criticar la práctica general de usar a los hijos de políticos como munición electoral.
Lo que quedó expuesto fue el uso de la inteligencia artificial no solo para distorsionar hechos o trayectorias políticas, sino para fabricar intimidades inexistentes y convertir la proximidad profesional en prueba de conspiraciones. Frente a eso, tres mujeres respondieron con una posición común: los menores no son instrumentos de campaña, y esa frontera debe mantenerse independientemente de la tecnología que se emplee para cruzarla.
A woman on X named Lucía Forero, appearing to support presidential candidate Abelardo de la Espriella, posted images generated by artificial intelligence that depicted two Colombian political figures—Paloma Valencia and Claudia López—in a supposed alliance. Alongside the fabricated pictures, Forero made a specific claim: that Angélica Lozano, López's wife, served as godmother to Amapola Rodríguez, Valencia's daughter. She framed this as evidence of hidden political ties, writing that the two candidates had deceived the public and that these were facts that deserved exposure. The post was later deleted, but not before the politicians named in it saw what had been written.
Valencia, a senator and opposition congresswoman, responded first and directly. She posted a message aimed at de la Espriella and his supporters with a single clear demand: do not involve her daughter in political matters. The language was sharp and personal—a mother drawing a line that she said should not be crossed.
López, the center-aligned presidential aspirant who positions herself against both far-right and far-left extremes, took a broader approach. She characterized the entire post as a weapon of disinformation, stating that lies and AI-generated fabrications were the tools being deployed against her. She denied both the alleged alliance with Valencia and the claim about Lozano being her daughter's godmother. López directed voters to her policy platform and criticized de la Espriella's campaign as desperate and crude, then made her own appeal: supporters should stop involving Valencia's child and any other minors in their electoral strategy.
Angélica Lozano, the senator and López's wife, entered the conversation to address the godmother claim directly. She noted that she had seen Valencia's daughter when the child was brought to Congress, just as she had encountered the children of other legislators from different political parties. The implication was clear—proximity to someone's child in a professional setting does not create a familial relationship. She questioned the logic of the accusation with a pointed rhetorical question: did seeing the children of various congresspeople make her godmother to all of them?
A political columnist named Sixto González asked Valencia publicly to clarify whether Lozano was indeed her daughter's godmother. Valencia's response was unambiguous: no. She called on González to spread her denial and stop sowing discord. Lozano responded again, this time criticizing the broader practice of involving politicians' children in electoral strategies at all. She reiterated that although she had seen Valencia's daughter at Congress, she held no familial role in the child's life.
What emerged from this exchange was a collision between two campaign strategies and two visions of what should be off-limits in politics. On one side, someone had weaponized artificial intelligence to create false visual evidence of an alliance and then layered onto it a personal claim designed to suggest hidden connections between political opponents. On the other side, three politicians—two of them targets of the disinformation—drew a sharp boundary around children, arguing that minors should not be conscripted into electoral warfare, regardless of the technology used to do so. The incident laid bare how AI-generated content could be deployed not just to mislead about policy or record, but to manufacture false intimacies and suggest conspiracies where none existed.
Citas Notables
With my daughter, do not interfere— Paloma Valencia, in response to the disinformation campaign
Their weapons are lies and AI-fabricated montages. Everything they say and show is false— Claudia López, rejecting both the alleged alliance and the godmother claim
La Conversación del Hearth Otra perspectiva de la historia
Why would someone fabricate this specific claim about a godmother relationship? What does that detail accomplish that a simple alliance claim wouldn't?
It's about plausibility and texture. An alliance between two politicians is abstract—people expect that. But a godmother relationship suggests something deeper, more personal, more hidden. It implies they've been concealing a real friendship, that their public positions are theater. It's the kind of detail that makes people think they've discovered something true.
And the fact that it involved a child—was that incidental or strategic?
Almost certainly strategic. Children are sympathetic figures. Involving a child makes the story more emotionally resonant and harder to dismiss. It also puts the politicians in a position where they have to defend their family, which shifts the conversation away from policy.
Why did Valencia and López respond so differently, then?
Valencia went straight to protection—don't touch my daughter. López went broader, attacking the entire apparatus of disinformation and the campaign using it. Valencia was defending her child; López was defending the integrity of the election itself.
Do you think the fact that Lozano had actually seen Valencia's daughter at Congress made this claim harder or easier to spread?
Easier. It gave the lie just enough grain of truth to seem plausible. Lozano had been around the child. That proximity could be twisted into something it wasn't. That's what made the denial so important—Lozano had to explain that seeing someone's child doesn't make you family.
What does this tell us about how AI disinformation works in electoral politics?
It shows that the technology isn't just about creating convincing fakes. It's about creating plausible narratives that exploit real relationships and real moments, then wrapping them in false visual evidence. The AI images were the hook, but the real lie was the story attached to them.