Artificial intelligence (AI) labs have begun to make their presence felt in the field of cybersecurity, primarily via application security. They have transitioned from static analysis to dynamic testing, leveraging their inherent code generation capabilities. However, industry trends suggest this is where the potential for disruption has its limits.

According to Sid Trivedi, a partner at Foundation Capital, the focus of many AI labs is predominantly horizontal. This means that they aim for widespread applicability and mainstream integration rather than delving deeply into specialized areas such as security. “Going deeply embedded into security doesn’t make a lot of sense for them,” Trivedi remarked, highlighting the inherent challenges of this paradigm.

Despite these limitations, as AI-generated code continues to rise in prominence, the need to secure that output has emerged as a natural progression for these labs. This development pushes them further into the realm of development-stage security, where the challenges and complexities of ensuring the safety of generated code become increasingly prominent.

In his assessment, Trivedi pinpointed three specific areas where he believes disruption from AI labs is unlikely to occur. The first involves runtime sensors, which necessitate an in-depth endpoint build-out. The second area concerns security functions built on proprietary data that cannot be openly trained upon. Finally, he noted the importance of Security Operations Center (SOC) and incident response workflows, which rely heavily on multi-tool integrations and embedded context to function effectively. Yet, amidst these challenges, Trivedi argues that identity management remains a crucial frontier ripe for exploration within the cybersecurity landscape.

Trivedi shared these insights during a video interview with the Information Security Media Group at the RSAC Conference 2026, where he further elaborated on the evolving landscape of cybersecurity and AI’s role within it. During the discussion, he brought attention to several aspects worth noting.

• He posited that Google currently holds the most robust end-to-end position among AI labs specializing in cybersecurity.
• He speculated about the possibility of AI labs eventually replicating the cloud service provider model, particularly concerning the management of identity at scale.
• Furthermore, he discussed how automating testing workflows represents an emerging expansion layer in the space beyond mere AI-driven code generation.

In his capacity at Foundation Capital, Trivedi focuses his investments on the enterprise stack, particularly within cybersecurity and IT infrastructure. His extensive background prior to joining the firm includes serving as an investment associate at Omidyar Technology Ventures, where he was actively involved in investing in and advising early-stage enterprise software companies. This breadth of experience provides him with a nuanced understanding of both the challenges and opportunities within the cybersecurity arena, particularly as it converges with rapidly evolving AI technologies.

As organizations navigate the complexities of integrating AI into their cybersecurity frameworks, Trivedi’s insights serve as a valuable resource for understanding the limitations and potential of AI labs in providing security solutions. The broader implications on the cybersecurity landscape will continue to unfold as these technologies mature and more enterprises explore their applications.