AI-Powered Fake News Fuels Bank Run Risks, UK Study Warns
The proliferation of artificial intelligence (AI)-generated fake news on social media platforms poses a significant and escalating threat to financial stability, increasing the risk of bank runs, according to a new study by UK research company Say No to Disinfo and communications firm Fenimore Harper. The study highlights the ease with which malicious actors can leverage generative AI to create and disseminate disinformation, including fabricated news stories and memes, that can erode public trust in financial institutions and trigger mass withdrawals. The speed and reach of social media amplify these risks, allowing disinformation to spread rapidly and potentially incite panic among depositors.
The findings of the study underscore the growing concerns among banks and regulators about the potential for social media-fueled bank runs, particularly in the wake of the Silicon Valley Bank collapse in 2023. During that event, depositors withdrew a staggering $42 billion within a mere 24 hours, demonstrating the destructive power of rapid information dissemination and subsequent panic. The rapid advancements in AI technology have compounded these risks, providing sophisticated tools for creating highly convincing fake news that can easily bypass traditional fact-checking mechanisms. The G20’s Financial Stability Board has also recognized this emerging threat, warning in November that generative AI could be weaponized by malicious actors to trigger financial crises, including flash crashes and bank runs.
The research involved presenting UK bank customers with samples of AI-generated fake news related to banking security and financial stability. The results were alarming: one-third of the participants indicated they were "extremely likely" to withdraw their money after seeing the fabricated content, while an additional 27% stated they were "somewhat likely" to do so. This high susceptibility to disinformation underscores the potential for even relatively small-scale disinformation campaigns to trigger significant financial instability. The speed with which individuals can now move their money via online and mobile banking further exacerbates the situation, allowing for near-instantaneous reactions to perceived threats.
The study estimates the potential financial impact of AI-powered disinformation campaigns to be substantial. For every £10 (approximately $12.48) spent on social media advertising to promote fake news, as much as £1 million in customer deposits could be withdrawn. This estimation is based on average deposit amounts held by UK customers, the cost of social media advertising, and projections of the reach and influence of such campaigns. The researchers emphasize the critical need for banks to implement robust monitoring systems that track both media and social media mentions, integrating these with withdrawal monitoring systems to detect and respond to any correlation between malicious information and changes in customer behavior.
The study’s findings have prompted calls for increased vigilance and proactive measures from financial institutions. Woody Malouf, head of financial crime at Revolut, a London-based fintech company, highlighted the importance of real-time monitoring for emerging threats, both within their customer base and across the wider financial ecosystem. He acknowledged the possibility of a significant industry event triggered by disinformation and stressed the need for preparedness among financial institutions. Malouf also emphasized the responsibility of social media platforms in curbing the spread of harmful content and mitigating these risks.
Despite the escalating concerns surrounding AI-generated disinformation, the overall sentiment within the banking industry regarding AI remains cautiously optimistic. Many institutions view AI as a powerful tool with the potential to enhance efficiency and improve customer service. They are actively working to manage and mitigate the risks associated with AI, while regulatory authorities focus on addressing the potential financial stability challenges posed by this evolving technology. The recent AI Summit in France, while emphasizing the positive aspects of AI development, reflects a shift in focus from the previous summit, which primarily addressed risk management. This evolving landscape requires a balanced approach that recognizes both the opportunities and the potential pitfalls of AI in the financial sector. The challenge lies in harnessing the power of AI while simultaneously safeguarding against its potential misuse for malicious purposes, ensuring the stability and integrity of the financial system.