AI2030 Responsible AI Tools Marketplace

Key Takeaways

Shadow AI use undermines institutional control: Financial institutions struggle with employees using public AI tools (like ChatGPT) for threat intelligence work outside official channels, creating security vulnerabilities and governance blind spots that prevent proper oversight of AI-driven security decisions.
AI tools are deployed before integration planning: Banks often adopt AI-based CTI solutions based on licensing availability rather than ensuring these tools actually fit into existing security workflows, leading to underutilization and reduced trust among security analysts who need seamless integration to adopt new technologies.
Security teams lack adversarial threat modeling: Financial institutions fail to adequately consider how attackers might exploit or deceive AI-driven threat intelligence systems, leaving a critical gap between how AI performs in controlled settings and how it performs against sophisticated, adaptive threats in real-world scenarios.
AI models themselves lack security safeguards: The research identifies a significant blind spot where financial institutions do not adequately monitor their AI systems for degradation, test their robustness against attacks, or maintain audit-ready documentation—creating hidden risks in the very tools meant to protect against cyber threats.
Trustworthy AI requires more than accuracy: Successful AI-driven cyber threat intelligence in finance depends on governance structures, workflow integration, and analyst trust as much as predictive performance, suggesting that technical improvements alone cannot solve adoption barriers without addressing organizational and human factors.

Abstract

Financial institutions face increasing cyber risk while operating under strict regulatory oversight. To manage this risk, they rely heavily on Cyber Threat Intelligence (CTI) to inform detection, response, and strategic security decisions. Artificial intelligence (AI) is widely suggested as a means to strengthen CTI. However, evidence of trustworthy production use in finance remains limited. Adoption depends not only on predictive performance, but also on governance, integration into security workflows and analyst trust. Thus, we examine how AI is used for CTI in practice within financial institutions and what barriers prevent trustworthy deployment. We report a mixed-methods, user-centric study combining a CTI-finance-focused systematic literature review, semi-structured interviews, and an exploratory survey. Our review screened 330 publications (2019-2025) and retained 12 finance-relevant studies for analysis; we further conducted six interviews and collected 14 survey responses from banks and consultancies. Across research and practice, we identify four recurrent socio-technical failure modes that hinder trustworthy AI-driven CTI: (i) shadow use of public AI tools outside institutional controls, (ii) license-first enablement without operational integration, (iii) attacker-perception gaps that limit adversarial threat modeling, and (iv) missing security for the AI models themselves, including limited monitoring, robustness evaluation and audit-ready evidence. Survey results provide additional insights: 71.4% of respondents expect AI to become central within five years, 57.1% report infrequent current use due to interpretability and assurance concerns and 28.6% report direct encounters with adversarial risks. Based on these findings, we derive three security-oriented operational safeguards for AI-enabled CTI deployments.

Key Takeaways

Shadow AI use undermines institutional control: Financial institutions struggle with employees using public AI tools (like ChatGPT) for threat intelligence work outside official channels, creating security vulnerabilities and governance blind spots that prevent proper oversight of AI-driven security decisions.
AI tools are deployed before integration planning: Banks often adopt AI-based CTI solutions based on licensing availability rather than ensuring these tools actually fit into existing security workflows, leading to underutilization and reduced trust among security analysts who need seamless integration to adopt new technologies.
Security teams lack adversarial threat modeling: Financial institutions fail to adequately consider how attackers might exploit or deceive AI-driven threat intelligence systems, leaving a critical gap between how AI performs in controlled settings and how it performs against sophisticated, adaptive threats in real-world scenarios.
AI models themselves lack security safeguards: The research identifies a significant blind spot where financial institutions do not adequately monitor their AI systems for degradation, test their robustness against attacks, or maintain audit-ready documentation—creating hidden risks in the very tools meant to protect against cyber threats.
Trustworthy AI requires more than accuracy: Successful AI-driven cyber threat intelligence in finance depends on governance structures, workflow integration, and analyst trust as much as predictive performance, suggesting that technical improvements alone cannot solve adoption barriers without addressing organizational and human factors.

Abstract

Security Barriers to Trustworthy AI-Driven Cyber Threat Intelligence in Finance: Evidence from Practitioners

Key Takeaways

Abstract

Paper Info

Responsible AI Topics

Links

Security Barriers to Trustworthy AI-Driven Cyber Threat Intelligence in Finance: Evidence from Practitioners

Key Takeaways

Abstract

Paper Info

Responsible AI Topics

Links