US and UK cybersecurity leaders are now as likely to attribute their company’s most significant security incident in the past year to employees’ unsanctioned use of artificial intelligence as they are to phishing attacks, according to a recent survey by the information security and compliance platform IO. This finding signals a dramatic shift in the threat landscape, with the rapid adoption of generative AI tools by employees creating new and complex security vulnerabilities that organizations are still struggling to fully comprehend and mitigate.

The comprehensive survey, which polled 3,000 cybersecurity managers and executives, revealed that among respondents who experienced a security incident within the last 12 months, approximately a quarter attributed their most serious incident or near miss to what is being termed "shadow AI." This figure is on par with the percentage blaming phishing or social engineering tactics. "Shadow AI" encompasses a range of risky behaviors, including employees granting AI tools unapproved access to sensitive business systems and inadvertently leaking confidential corporate information through these platforms. This trend highlights a critical gap in organizational control and employee awareness regarding the secure and compliant use of emerging AI technologies.

While phishing has historically been the primary focus of cybersecurity awareness training programs within organizations, remaining the most commonly experienced type of incident at 32%, the data indicates that shadow AI is rapidly challenging its dominance when it comes to the most severe security breaches. The survey further found that using generative AI without explicit organizational permission was identified as the most prevalent employee security mistake, cited by 35% of respondents. Adding to this concern, 22% of participants reported that staff members have integrated AI into crucial business workflows without the knowledge or oversight of the security team. This indicates a widespread "bring your own AI" (BYOAI) phenomenon, where employees leverage readily available AI tools for productivity gains without considering the associated security implications, creating blind spots for IT and security departments.

The implications of this shift are profound. Organizations must not only reinforce traditional security measures but also develop robust strategies to manage and monitor the use of AI. This includes establishing clear policies on acceptable AI usage, implementing technical controls to detect and block unauthorized AI access, and conducting targeted training programs that educate employees on the specific risks associated with generative AI, such as data leakage, intellectual property theft, and the potential for AI-generated misinformation to be used in further attacks. The rise of shadow AI suggests that current security awareness initiatives may not be adequately addressing the nuances of these new technologies.

Legal Leaders Grapple with AI Governance Lag

The challenges posed by the rapid integration of AI are not confined to cybersecurity alone. A separate survey conducted by Diligent Institute in collaboration with law firm Wilson Sonsini reveals a significant disconnect between the pace of AI adoption and the development of corresponding governance frameworks. More than half of legal leaders surveyed believe that their organizations are embracing AI at a rate that outstrips their ability to govern the technology effectively.

This sentiment was echoed by 59% of the approximately 200 legal leaders surveyed, who indicated that organizational governance is struggling to keep pace with the swift adoption of AI technologies. The findings suggest a widespread lack of established protocols for AI implementation and oversight. Specifically, over two in five respondents reported that their organizations lack a basic AI governance framework. Digging deeper, 14% admitted to having no formal approach whatsoever, while an additional 27% are still in the process of developing their policies and guidelines. On a more positive note, 38% of organizations have a basic framework in place, and 21% reported that their AI governance structures are well-established and undergo regular review.

This governance gap emerges at a time when organizations are increasingly apprehensive about AI-related threats. Nearly half of the legal leaders surveyed (49%) identified AI-related risks as among the leading concerns for their organizations. This represents a notable increase of 10 percentage points compared to an earlier survey conducted in 2026. While AI risks are escalating, the changing regulatory environment was cited as the top risk factor by 55% of respondents, an increase of seven percentage points. In contrast, concerns related to geopolitical conflict have sharply declined as a risk factor, dropping from 52% to 26% in the same period. This shift indicates a growing focus on the immediate and evolving legal and compliance landscapes surrounding AI, driven by the rapid development and deployment of these technologies.

The lack of comprehensive AI governance frameworks presents a critical challenge for businesses. Without clear guidelines, organizations risk non-compliance with emerging regulations, potential legal liabilities, reputational damage, and an inability to effectively manage the ethical implications of AI deployment. This necessitates a proactive approach from legal and compliance departments to develop and implement robust governance structures that can adapt to the dynamic nature of AI technology and its associated risks.

The Escalating Cost of Data Breaches

Adding to the complex landscape of corporate risk, the financial implications of data breaches continue to be a significant concern. A comprehensive study by Cornerstone Research has revealed that data breach settlements in federal class-action lawsuits have cumulatively amounted to $3.2 billion between 2010 and 2025. This period has witnessed a dramatic and sustained increase in the frequency and financial scale of such settlements, underscoring the growing vulnerability of organizations to data security failures and the escalating legal and financial repercussions.

The study meticulously examined 426 federal class-action settlements related to data breaches over the 15-year period. Of these, 357 provided disclosed details regarding settlement amounts. The findings paint a stark picture of escalating costs and an increasing number of legal actions. The financial services, technology, and healthcare industries have borne the brunt of these settlements, indicating that sectors handling vast amounts of sensitive personal and financial information are prime targets and face the most significant financial exposure.

To illustrate the trajectory of these costs, the single data breach settlement with a known amount in 2010 was $13 million. For the first half of the decade (2010-2015), the number of data breach settlements remained relatively low, with no more than six occurring annually, and none exceeding $50 million. However, a significant inflection point occurred from 2016 onwards. The number of settlements began a steep ascent, reaching 36 in 2022, surging to 51 in 2023, and then escalating dramatically to 91 in 2024 and a peak of 139 in 2025.

The financial impact mirrored this surge in litigation. Settlement dollars also increased substantially, reaching a peak in 2020 with a staggering $889 million. This figure was largely driven by a single, substantial settlement of $629 million in a lawsuit against Equifax, a major credit reporting agency, following a massive data breach. While the occurrence of very high-value settlements, defined as $50 million or more, has seen a decline, with none recorded in 2025, the overall settlement volume remained high, contributing to a total of $467 million in that year.

A granular analysis of settlements with monetary compensation reveals that the majority, 76%, were valued at $5 million or less. A further 18% fell within the $5 million to $25 million range, with 3% each in the $25 million to $50 million and $50 million or more categories. This distribution highlights that while a few colossal settlements can significantly skew the total figures, a vast number of smaller to moderate settlements also contribute substantially to the overall financial burden on affected organizations.

Breaking down the industry impact, financial services emerged as the most heavily impacted sector, having paid almost $1.2 billion in data breach settlements between 2010 and 2025. The technology industry followed with $726 million, and the healthcare sector ranked third, incurring $585 million in settlements. Interestingly, despite the significant financial toll, the healthcare industry recorded the highest number of settlements overall, with 118 cases. This suggests that while healthcare organizations face a high volume of litigation, the average settlement value might be lower compared to other industries.

The consistent rise in data breach settlements underscores the critical need for organizations to prioritize robust data security measures. This includes investing in advanced cybersecurity technologies, implementing comprehensive data protection policies, conducting regular security audits, and fostering a culture of security awareness among all employees. Furthermore, organizations must be prepared for increased regulatory scrutiny and potential litigation, necessitating strong legal and compliance frameworks to manage the fallout from potential breaches. The escalating costs serve as a stark reminder of the financial and reputational damage that can result from inadequate data security.

Interconnected Risks and the Path Forward

The confluence of these findings – the rise of shadow AI, the governance lag in AI adoption, and the persistent and escalating costs of data breaches – paints a complex picture of the modern risk landscape. Cybersecurity, governance, and legal departments are increasingly intertwined, facing a shared challenge in navigating the rapid evolution of technology and its implications.

The unchecked use of generative AI by employees creates immediate security vulnerabilities, as highlighted by the IO survey. These vulnerabilities can lead to data leakage, intellectual property compromise, and the potential for sophisticated phishing or social engineering attacks powered by AI. Simultaneously, the Diligent Institute’s findings reveal that organizations are often ill-equipped to govern these powerful new tools, lacking the necessary frameworks to establish clear policies, monitor usage, and ensure compliance. This governance gap exacerbates the security risks, creating an environment where employees may inadvertently or intentionally expose sensitive information.

The historical data on data breach settlements from Cornerstone Research provides a long-term perspective on the financial consequences of security failures. The multi-billion dollar figures demonstrate that breaches are not merely operational disruptions but can lead to substantial financial liabilities that impact an organization’s bottom line for years. As AI introduces new vectors for breaches, the potential for future settlement costs could escalate further if not proactively managed.

The implications are clear: organizations must adopt a holistic and integrated approach to risk management. This involves:

  • Enhanced Employee Training: Security awareness programs need to be updated to specifically address the risks of generative AI, including data privacy, prompt injection, and the ethical considerations of AI use.
  • Robust AI Governance Frameworks: Legal, compliance, and IT departments must collaborate to establish clear policies on AI adoption, usage, and oversight. This includes defining acceptable use cases, implementing data access controls, and conducting regular risk assessments of AI deployments.
  • Proactive Security Measures: Organizations need to invest in advanced security technologies capable of detecting and mitigating AI-driven threats. This could include AI-powered threat detection systems and data loss prevention (DLP) solutions that can monitor AI tool usage.
  • Cross-Departmental Collaboration: Breaking down silos between cybersecurity, legal, and IT departments is crucial. A unified approach to understanding and managing emerging risks is essential for effective risk mitigation.
  • Regulatory Vigilance: Staying abreast of the evolving regulatory landscape surrounding AI and data privacy is paramount. Proactive engagement with regulatory bodies and ensuring compliance with new mandates will be critical.

The trend of employees leveraging powerful new technologies without full organizational awareness or control is a double-edged sword. While it can drive innovation and productivity, it also introduces significant risks. By understanding these interconnected challenges and implementing comprehensive, integrated strategies, organizations can better navigate the complexities of the modern digital landscape and safeguard themselves against escalating threats. The future of corporate risk management will undoubtedly hinge on an organization’s ability to effectively manage the dual imperatives of technological adoption and robust governance.

By