The Rising Security Challenges of AI: Insights from Recent Survey of CISOs and CTOs

In the rapidly evolving landscape of technological advancements, artificial intelligence (AI) continues to play a pivotal role in reshaping business operations and security protocols. A recent report from NetFoundry sheds light on the growing concerns surrounding AI security, revealing critical insights from 200 Chief Information Security Officers (CISOs) and Chief Technology Officers (CTOs) in various industries.

Key Findings of the Report

The report, titled 2026 State of Secure AI Access, shows that 100% of surveyed leaders acknowledge that AI is amplifying their organization's attack surface. This is an alarming statistic, particularly as organizations anticipate an additional 14% increase in their attack surface over the next year. The heightened risk landscape posed by AI does not come at a time of abundance in confidence regarding security measures. Only 15% of surveyed leaders express a high level of confidence in their existing tools to adequately protect AI frameworks.

Moreover, when it comes to securing machine-to-machine connectivity, there appears to be a significant oversight, with a staggering 85% of respondents actively seeking innovative methods to safeguard non-human identities. The report suggests that organizations are facing a dual challenge: the pressure to deploy AI responsibly and the realization of inadequacies in their current security infrastructures.

The Pressure Cooker Scenario

Interestingly, while 78% of respondents reported feeling substantial pressure to secure AI implementations, their unease is palpable. A notable 93% voiced concerns about the new risks introduced by AI, with more than half of them being highly apprehensive. The juxtaposition of immediate operational demands against security uncertainties is creating a daunting scenario for technical leaders.

CISOs emerged as particularly cautious, with only 10% expressing confidence in their ability to protect AI-driven initiatives, compared to 18% of their CTO counterparts—indicating a noticeable gap in comfort levels concerning AI security.

A Shift in Security Paradigm

Traditionally, security has revolved around human identities. NetFoundry’s CEO, Galeal Zino, aptly noted, “AI has fundamentally changed what enterprises need to secure.” The survey illuminates a crucial pivot towards an identity-first approach that prioritizes securing machine interactions over the human aspect of connectivity. The focus is not only on the tools but also on rethinking strategies to encompass the complexities of machine identities that are crucial to sustaining organizational security.

The Broader Implications

As AI adoption surges, the Gartner® 2025 Machine Identity study highlights the pressing need for formal machine identity and access management strategies. A staggering 42% of organizations lack these frameworks, exacerbating their vulnerability and risking security events arising from unmanaged machine identities. This gap in an organization’s security architecture poses a critical challenge that needs immediate attention.

The data indicates that while companies are keen to leverage AI, there's a pressing need to recalibrate their security strategies. With 58% of organizations reporting security incidents due to the neglect of machine identity protocols, it's clear that enhancing visibility and control over machine interactions is paramount for current and future digital infrastructures.

Conclusion

The findings from NetFoundry's survey serve as a clarion call for organizations to reassess their security frameworks in the wake of AI proliferation. As they navigate this complex terrain, the emphasis on creating a robust, identity-centric approach to machine connectivity is vital. This approach not only aims to minimize the attack surface but also strives to fortify defenses against evolving digital threats. By shedding light on these crucial insights, the report paves the way for a more secure implementation of AI technologies across sectors.