From Humans to Machines: The Evolution of Identity in the AI Era

Read full article here: https://www.andromedasecurity.com/videos/cyberbytes-interview-murali-basavaiah-identity-access-ai/?utm_source=nhimg

In this episode of CyberBytes, Murali Basavaiah, co-founder and CEO of Andromeda, explores one of the most urgent and misunderstood challenges in modern cybersecurity — identity security in the era of artificial intelligence. Drawing from his journey as a serial entrepreneur, Murali unpacks the transformation of identity security from traditional access control to AI-driven, prevention-focused protection, and explains how agentic AI is reshaping both the threat landscape and the defense playbook.

The Growing Identity Security Crisis

Identity is now the primary attack vector in cybersecurity. According to Murali, identity compromises account for roughly 80% of all breaches, a figure that reflects how deeply organizations depend on identities — both human and non-human — to function. With the rise of agentic AI, automation, and service accounts, enterprises are facing an explosion of non-human identities, growing by as much as 10x to 30x in complex cloud environments.

This surge in identities, coupled with legacy IAM tools and fragmented controls, has created a blind spot where dormant, over-permissioned, and unmonitored identities pose systemic risks. Traditional security approaches designed for human-centric environments are no longer enough to handle the scale, velocity, and autonomy of machine-driven identities.

AI-Powered Context-Aware Security: The Andromeda Approach

Andromeda’s platform uses AI-driven behavioral context modeling to transform how enterprises secure identities. By analyzing vast telemetry and audit log data across multi-cloud applications and infrastructures, Andromeda builds context-aware models that detect risky behaviors before they escalate into breaches.

This prevention-first philosophy marks a shift from reactive incident response to proactive identity protection. Murali emphasizes that the business impact of a breach correlates directly with entitlements, many of which are unused but remain highly privileged. AI helps identify and remove those dormant permissions automatically — drastically reducing exposure without affecting productivity.

The Impact of Agentic AI on Attack and Defense

The emergence of agentic AI is changing cybersecurity dynamics. These autonomous AI systems can now act independently, making decisions, running code, and interacting with other systems. Murali warns that agentic AI lowers the barriers to compromise, allowing attackers to execute sophisticated, multi-layered campaigns with less effort.

At the same time, the same AI capabilities are enabling defenders to analyze billions of identity signals in real time, contextualize anomalies, and automate remediation. The race between attacker automation and defensive intelligence is accelerating — and enterprises must keep pace by operationalizing AI-driven identity defense at scale.

From Dashboards to Conversations: The UX Evolution in Security

Murali highlights a major user experience shift in cybersecurity products. The traditional dashboard model — dense with metrics and manual navigation — is being replaced by conversational, AI-driven interfaces. These chat-based models allow analysts and engineers to query systems in natural language, instantly surfacing insights that previously took hours to extract.

This democratization of security intelligence empowers teams across skill levels to interact with complex data more intuitively, reducing friction and accelerating response times. However, Murali stresses that explainability is just as critical as innovation. Without clear reasoning behind AI-driven decisions, trust and adoption will remain limited.

Balancing Security and Productivity in the Enterprise

One of the enduring challenges for CISOs is balancing entitlement management — granting enough access for teams to remain productive without creating over-permissioned risk. In modern environments, that equilibrium is constantly shifting due to automation, role changes, and AI integrations.

Murali explains that the key lies in continuous, AI-driven entitlement analysis that dynamically adjusts access rights in response to behavior, not static policies. This real-time governance model helps enterprises minimize privilege sprawl while avoiding operational bottlenecks.

The Business and Operational Challenge of AI Security

While building AI-powered products has become easier, operationalizing AI security at enterprise scale remains a formidable challenge. Murali reflects that startups can now develop advanced solutions faster than ever, but go-to-market execution, customer education, and integration into existing workflows are the true differentiators.

For AI-driven identity security to succeed, solutions must seamlessly embed into enterprise ecosystems, provide actionable insights, and demonstrate measurable business value. Otherwise, even the most advanced AI models risk being underutilized or sidelined.

Future Outlook: AI-Driven Entitlements and Non-Human Identity Governance

As agentic AI becomes more autonomous, enterprises must start treating AI identities as first-class citizens within their IAM programs. Defining and governing the permissions, access scope, and entitlements of AI systems will become central to future identity architectures.

Murali predicts that the next wave of cybersecurity innovation will focus on AI-to-AI trust frameworks, where explainable models, behavioral baselines, and autonomous governance mechanisms prevent one AI system from compromising another.

Conclusion

“Identity in the Age of AI” captures a pivotal transformation in cybersecurity. As the number of machine and AI-driven identities skyrockets, organizations must move beyond legacy IAM and adopt context-aware, AI-powered, prevention-focused identity security. The future of cybersecurity depends on explainable, operationally viable AI systems that balance risk, trust, and productivity in an increasingly autonomous enterprise landscape.

Andromeda is positioning itself at the forefront of this evolution — leveraging artificial intelligence not just as a defensive mechanism, but as the strategic backbone for securing identities in the age of agentic AI.