How Is ‘AI Slop’ Directly Funding The Rise Of Cyber Fraud?

Cybersecurity: The Role of AI in Modern Threats

AI technology has become a significant factor in online security threats. Its capabilities enable criminals to automate scams, create fake content, and conduct social engineering attacks more efficiently.

According to research by the World Economic Forum's Global Cybersecurity Outlook, cyber fraud is prevalent, with 73% of respondents reportedly experiencing such incidents in 2025. Corporate leaders identify it as the top cybersecurity threat, with 77% reporting an increase in incidents.

Generative AI tools facilitate the creation of convincing messages and media rapidly. Criminals use these tools to produce emails, social posts, and advertisements that appear legitimate, as noted by the World Economic Forum. AI is described as enabling rapid content creation that allows criminals to scale and personalize scams efficiently.

These tools also contribute to automated advertising fraud systems. Campaigns like AutoBait demonstrate how generative AI can generate clickbait articles and images at a cost of approximately $2.25 per page. These pages, often placed on abandoned but previously legitimate websites, help attract traffic and bypass automated checks.

This method results in numerous advertising impressions, channeling advertising revenue to websites designed solely for such purposes.

Impact on Global Advertising Expenditures

Digital advertising fraud significantly impacts the global economy. Research indicates that $84 billion of global digital ad spending was lost to fraud in 2023, accounting for up to 22% of total online ad expenditures.

The scale of the problem is highlighted in the Digital Ad Fraud 101 document, which compiles various research sources. Estimates vary, but most studies place the cost in the tens of billions annually. Projections suggest that losses could exceed $170 billion each year by 2028 if current trends persist. Mobile advertising is particularly vulnerable, with approximately 30% of spending potentially affected by fraud.

Researchers have identified four primary destinations for fraudulent advertising revenue:

Organized cybercrime networks using botnets or malware-infected devices for fake clicks and ad impressions. Websites created specifically to harvest programmatic advertising revenue through low-quality or scraped content. Mobile fraud ecosystems using fake app downloads, click injection, and spoofed user behavior. Complex intermediary structures in advertising supply chains that absorb revenue through hidden fees and arbitrage.

Weak transparency in programmatic advertising complicates tracing these activities and allows bad actors to divert revenue or hide fraudulent traffic within legitimate flows.

AI tools have enabled criminals to create scalable scams. Technologies like synthetic voice can imitate a person's speech patterns, which are used in phishing calls or video meetings. A notable instance involved a British engineering company losing $25 million due to an AI-generated voice during a video call, convincing employees that the request was from a trusted executive.

Deepfake videos have also been used in political and social contexts. Examples include a fabricated clip of an Irish presidential candidate falsely announcing withdrawal from an election and a manipulated Instagram video falsely showing Indonesia’s president directing citizens to a WhatsApp number for financial aid.

The World Economic Forum states that fraud is a pervasive cyber risk affecting households, corporations, and national economies. It highlights how a single scam email can lead to data breaches, disrupting company operations and impacting supply chains and cross-border trust in digital systems.

Challenges in Combating AI-Driven Cybercrime

AI enhances both cyber defense tools and criminal tactics. Machine learning systems are employed to detect suspicious transactions in banking and payment systems. However, the same technology lowers the cost of deception, enabling criminals to generate convincing messages, images, and voices at low cost.

The World Economic Forum notes that "AI's potential to automate cybercrime may only be matched by its capacity to prevent it." The Global Cybersecurity Outlook identifies structural issues that weaken defenses, including varying regulations between countries, limited intelligence sharing, and a lack of cybersecurity skills in many organizations. SMEs are particularly affected, with 46% reporting critical skills shortages.

International discussions, such as those at the United Nations and Interpol Global Fraud Summit in Vienna, are bringing together governments, technology groups, and civil society to address cybercrime across borders.

Based on reporting by techround.co.uk.