In the age of generator AI, misinformation has grown beyond mere falsehood. It's about storytelling. Viral posting and theory of conspiracy not only involves inaccuracy, but also spreads to resonate emotionally with the audience.
Recent research shows that while AI can help detect these manipulative narratives, human judgment and emotional literacy continue to be important for understanding and rebutting the digital spread of misleading content.
AI Author
A study by Zhao, Zhou, and Wang tested 1,600 participants in three experiments to measure how AI creator labels and content types influence reader responses.
They found that when news was labelled as AI-generated, they always received lower cognitive and emotional ratings, regardless of their de facto accuracy.
However, emotionally charged content, including conspiracy and partisan news, elicited stronger engagement and sharing intentions, even when attributed to AI.
The author stated that “emotional resonance can negate source skepticism. […]there are new challenges, especially in suppressing content generated by emotionally manipulative AI.”
This shows the important point that emotions can negate skepticism about AI authors. In other words, simple labeling is not enough to stop the spread of the virus in the manipulation content.
Source Attribution
Luttrell, Davis, and Welch investigated the attribution of sources of AI-ERA journalism and found that traditional methods are increasingly inadequate. Text generated in AI is persuasive as automated systems often fail to detect it because they can mimic human writing.
Their research highlights that a single technical revision cannot ensure journalism against AI-era disinformation. Robust defense requires a layered strategy that combines detection, origin validation, and editorial judgments within the loop.
The author emphasizes the implementation of Content Consistency Theory (TOCC) as a framework for AI developers. “Design a more subtle set of assessments based on journalistic values.”
This highlights the need for human editorial monitoring, along with AI detection to identify subtle operations.
Journalist's perspective
A survey of 504 journalists in the Basque country by Peña-Alonso, Peña-Fernandez and Meso-Ayerdi found that almost 90% believe that AI will significantly increase the risk of disinformation.
Research shows that experienced journalists are aware of higher risk than their less experienced colleagues, and that journalists who frequently use AI tools have a more nuanced perception and are aware of both potential benefits and risks.
The authors conclude that concerns about AI-driven disinformation are almost unanimous, familiar with the perception of AI impact and emphasize the importance of professional literacy and training.
Journalists highlighted the challenges that included deepfake detection, forged datasets and content that fused facts with manipulative narratives. This is a reminder that human expertise remains essential.
Weaponized Storytelling
Research at Florida International University shows how AI can detect “weaponized storytelling.” There, disinformation manipulates the audience using personacues, culturally loaded symbols, and sequences of narratives.
AI tools analyze usernames and handles to infer reliability or partnerships, detect post sequences to detect manipulated timelines, and culturally specific symbols to reveal subtle patterns of impact that traditional fact-checking may overlook.
This shows that disinformation is not just about false facts, but also about how stories are told to evoke trust, fear, or other emotions.
The author explains his point in the following sentence: “The woman in a white dress was filled with joy.”
While you are in the west, this phrase evokes a happy image. In parts of Asia, white has the opposite meaning as it symbolizes death and mourning.
This study found that AI training on diverse cultural narratives increases sensitivity to such nuances.
Transparency and literacy
These studies present consistent themes. AI is powerful, but it cannot replace human judgment.
AI scans thousands of posts, detects stylistic markers, and detects flag inconsistencies, but humans need to interpret emotional nuances, cultural contexts, and ethical considerations.
There are also restrictions on binary AI labels. Emotional content promotes emotional and behavioral responses more than author clues, and inadequate brief disclosure.
Platforms, policymakers, and journalists need to combine AI detection with human analysis, editorial judgment, and public digital literacy.
Understanding the interactions of AI, emotions and narratives is important in a digital ecosystem where stories move faster than validation.
Fact checking alone is not enough. Fighting disinformation requires a multilayered approach that combines technical tools and human expertise with emotional intelligence and emotional intelligence, risking the agenda to manipulation rather than fact.
(BM)
