Targeted Disinformation
Table of Contents
A Story of Targeted Disinformation: The Case of Jane Smith
Jane Smith, a mid-level manager at a tech company in New York, was an active social media user. She enjoyed sharing her thoughts on current events, engaging in discussions about technology, and staying connected with friends and family. Unbeknownst to her, Jane had become the target of a sophisticated disinformation campaign orchestrated by a shadowy group focused on influencing public opinion, and executed entirely by AI under their control.
The Background
The campaign was driven by two advanced AI systems, working independently but in tandem, developed to influence public opinion on a controversial new piece of legislation affecting the tech industry. These AI systems were capable of executing the entire disinformation strategy independently, targeting millions of people simultaneously.
The first AI system focused on building detailed profiles of individuals like Jane. It analyzed her favorite news sources, her stance on various political issues, and even her daily routine. With this information, the AI crafted a highly personalized disinformation strategy designed to manipulate her views and draw her into an echo chamber.
The second AI system was responsible for the broader disinformation campaign. It utilized bots to promote the same misleading message across various social media platforms, amplifying the disinformation and reinforcing the narrative created by the first AI.
The Initial Hook
One morning, Jane received a message on her favorite social media platform. It appeared to be from a trusted friend, sharing a news article about the legislation. The article, which looked legitimate, contained a mix of true and fabricated information. It suggested that the legislation was secretly designed to benefit a few powerful tech companies at the expense of smaller businesses and employees like Jane.
The article included quotes from supposed experts and links to other fake articles and manipulated videos supporting its claims. One video featured a deep fake of a well-known tech industry leader making disparaging remarks about employees and small businesses. The content was highly persuasive, playing on Jane’s existing concerns and biases.
Drawing Her In
Convinced by the initial article, Jane started engaging with it by liking, commenting, and sharing it on her feed. This engagement triggered the various other social media AI’s algorithms, which now targeted her with even more tailored content. Over the next few days, Jane’s social media feeds were flooded with similar posts, comments, and videos. These posts were shared by seemingly real accounts, including some that she followed, further lending credibility to the false narrative.
As Jane interacted more with this personalized content, her feeds became almost fully those developed and promoted by the second AI system. This one is focused on crafting mass disinformation campaigns with creating lots of semi-fake, manipulative content and using bots to promote the broader disinformation campaign, ensuring that Jane and others like her were continuously exposed to the same misleading messages across various platforms. The content she saw became increasingly extensive, encompassing more aspects of the legislation and its supposed impacts. Articles and videos began to surface, depicting false statistics, fabricated testimonies, and sensationalist headlines that all pointed to a grand conspiracy against small tech businesses.
The Echo Chamber
By now, Jane has been drawn into the conspiracy with the targeted disinformation AI, and has been increasingly consuming the material developed by the broad disinformation campaign which abuses social media’s own AI systems to increase her exposure to the manipulative content.
She found herself drawn into an echo chamber of disinformation. She now joined groups and forums where like-minded individuals, also targeted by the AI, discussed and amplified the disinformation. These groups created a sense of community and shared purpose, further entrenching Jane’s beliefs.
Continously, the second AI system’s broader campaign ensured that these echo chambers were filled with consistent, reinforcing disinformation. The bots continued to promote and amplify the false narratives, making them appear more credible and widespread.
Jane began to change her views significantly. She started discussing the issue with her colleagues and friends, sharing the fake articles and videos she had seen. Her passionate arguments and well-researched (but false) information swayed some of her peers, spreading the disinformation further. Her posts gained traction, catching the attention of local media and even a few policymakers. The dual AI-driven disinformation campaign had successfully amplified its reach, influencing not just Jane and her immediate circle but also a broader audience.
The Aftermath
Eventually, a few diligent journalists and fact-checkers uncovered the disinformation campaign. They traced the origins of the fake news articles and deep fake videos, exposing the advanced AI systems behind them. Major news outlets reported on the incident, and social media platforms took down the offending content.
However, when Jane learned about the exposure of the disinformation campaign, she didn’t see it as proof of her manipulation. Like for most people, Jane’s ego didn’t allow her to admit to herself that she was wrong and that she was so easily manipulated. Instead, she convinced herself that the exposure was just another layer of conspiracy. She felt increasingly isolated and defensive, clinging to the belief that there was a grand scheme against people like her.
Jane doubled down on her convictions. She sought out even more like-minded individuals in fringe groups and forums, where her views were validated and amplified. Her resentment towards the legislation, which she, wrongly, believed was designed to harm her and her peers, grew stronger. She became convinced that a powerful cabal controlled all the information that contradicted her beliefs.
With none of the AI systems in the picture anymore, Jane continued being impacted by the disinformation campaign. Her interactions with friends, family, and colleagues became strained as her new worldview clashed with reality. Jane’s life took a downward spiral, as she became more entangled in the web of disinformation. Her entire perspective shifted, and she found herself increasingly isolated from the mainstream, living in a world dominated by conspiracy theories and mistrust.
Jane’s experience highlights the dangers of targeted disinformation and the sophisticated techniques used to draw individuals into broader disinformation campaigns. It shows how advanced AI tools can create highly personalized and persuasive false information, exploiting individuals’ beliefs and biases to suck them into an echo chamber.
Targeted Disinformation: Understanding the Threat
Among the various forms of disinformation, targeted disinformation stands out as particularly insidious and effective. As illustrated by Jane’s story above.
What is Targeted Disinformation?
Targeted disinformation refers to the deliberate spread of false or misleading information tailored to specific individuals or groups based on their personal characteristics, beliefs, and behaviors. Unlike broad-spectrum disinformation, which aims to influence a wide audience, targeted disinformation is highly personalized. It leverages detailed data about its targets to craft messages that are more likely to resonate with them and achieve the desired manipulative effect.
This form of disinformation can take many shapes, including fake news articles, doctored images, manipulated videos (such as deep fakes), and even falsified personal communications. The key characteristic is that the content is designed to appeal specifically to the recipient’s preferences, fears, and biases, thereby increasing its persuasive power.
Why is Targeted Disinformation Particularly Risky?
The risks associated with targeted disinformation are multifaceted and profound. By tailoring false information to closely match the beliefs and biases of individuals, targeted disinformation can significantly erode trust in traditional information sources, including the media, public institutions, and even personal relationships. The personalized nature of this disinformation makes it more persuasive than general falsehoods, as it exploits the recipient’s existing views and emotions, making the false information more believable and likely to be acted upon.
Targeted disinformation also has the potential to deepen societal divides and exacerbate conflicts. By targeting specific groups with disinformation that reinforces their prejudices or fears, perpetrators can create and amplify social division. This can lead to increased polarization and even social unrest, as seen in numerous incidents where disinformation has sparked protests or violence.
In the context of democratic processes, targeted disinformation can be particularly dangerous. It can be used to influence elections and political decisions by swaying voters’ opinions in a highly strategic manner. This undermines the democratic process and can lead to illegitimate outcomes, as the electorate is manipulated by false information crafted to exploit their specific beliefs and emotions.
The psychological impact on individuals targeted by disinformation is also significant. As seen in the story of Jane Smith, victims of targeted disinformation may become entrenched in echo chambers that validate and reinforce their manipulated beliefs. This can lead to increased isolation, resentment, and a distorted worldview, severely impacting their personal and professional lives.
How AI Enables Targeted Disinformation
Artificial intelligence plays a crucial role in the creation and dissemination of targeted disinformation. Advanced AI tools can analyze vast amounts of data from social media, browsing histories, and other digital footprints to build detailed profiles of individuals. These profiles include information about users’ interests, political affiliations, emotional states, and social connections, which are then used to craft personalized disinformation campaigns.
AI-powered tools can generate realistic and persuasive disinformation content. For instance, models can create fake news articles or social media posts that mimic the writing style and tone of legitimate sources. Additionally, deep learning algorithms can produce deep fakes—videos and audio recordings that convincingly depict individuals saying or doing things they never did.
Using techniques borrowed from digital advertising, AI can deliver disinformation to specific individuals or groups with precision. By leveraging data on users’ online behaviors and preferences, disinformation campaigns can ensure that false messages reach the most susceptible targets at the most opportune times. This micro-targeting significantly enhances the effectiveness of disinformation efforts.
Furthermore, AI-driven bots and automated accounts can amplify disinformation by sharing and promoting false content across various platforms. These bots can interact with real users, making the disinformation appear more popular and credible, thus increasing its reach and impact. AI systems can continuously learn from the success or failure of disinformation campaigns, refining and optimizing future efforts to make them increasingly effective.
Combating Targeted Disinformation
Addressing the threat of targeted disinformation requires a comprehensive approach that combines technological, regulatory, and educational strategies. Social media platforms must enhance their algorithms to better detect and de-emphasize disinformation. This involves leveraging advanced machine learning techniques to recognize the linguistic and contextual markers of false information and ensuring that these models are continuously updated to stay ahead of evolving tactics.
Regular audits of AI systems are essential to ensure that they operate fairly and transparently. Algorithmic auditing involves examining the data, processes, and outcomes of AI systems to identify and correct biases that may have been inadvertently introduced. This process helps maintain the integrity and reliability of AI systems used in content moderation and disinformation detection.
Developing technological solutions to detect and authenticate content is critical in combating deep fakes and other forms of manipulated media. Forensic tools that analyze video and audio files for signs of manipulation can detect anomalies indicating tampering. Additionally, digital provenance solutions, which involve watermarking content at the time of creation, can help verify the authenticity of media files.
Crafting regulatory frameworks that balance the need for content moderation with the protection of free speech is a delicate but necessary task. Policymakers must develop laws and policies that address the misuse of AI in disinformation while safeguarding fundamental rights. Co-regulation, involving collaboration between governments, tech companies, and civil society, offers a balanced approach to tackling disinformation.
Public awareness campaigns and media literacy programs are vital in equipping individuals with the skills to identify and critically evaluate disinformation. Educating the public about the existence and potential impact of AI-generated disinformation can help build a more informed and resilient society.
Conclusion
Targeted disinformation poses a significant threat to societal trust, democratic processes, and individual well-being. The use of AI in these disinformation campaigns enhances their precision, persuasiveness, and impact, making them more dangerous than ever before. By understanding the mechanisms of targeted disinformation and implementing comprehensive strategies to combat it, society can better protect itself against these sophisticated threats.
For 30+ years, I've been committed to protecting people, businesses, and the environment from the physical harm caused by cyber-kinetic threats, blending cybersecurity strategies and resilience and safety measures. Lately, my worries have grown due to the rapid, complex advancements in Artificial Intelligence (AI). Having observed AI's progression for two decades and penned a book on its future, I see it as a unique and escalating threat, especially when applied to military systems, disinformation, or integrated into critical infrastructure like 5G networks or smart grids. More about me, and about Defence.AI.
Luka Ivezic
Luka Ivezic is the Lead Cybersecurity Consultant for Europe at the Information Security Forum (ISF), a leading global, independent, and not-for-profit organisation dedicated to cybersecurity and risk management. Before joining ISF, Luka served as a cybersecurity consultant and manager at PwC and Deloitte. His journey in the field began as an independent researcher focused on cyber and geopolitical implications of emerging technologies such as AI, IoT, 5G. He co-authored with Marin the book "The Future of Leadership in the Age of AI". Luka holds a Master's degree from King's College London's Department of War Studies, where he specialized in the disinformation risks posed by AI.