Growth Drivers | October 14, 2025

Securing the machines behind the machines

By Martin Reilly, Digital Identity Offering Manager at DXC Technology


We live in an increasingly connected world — from wearables on our wrists and smart devices in our homes to connected vehicles and intelligent infrastructure that power our cities and keep our societies moving.   
 
Our physical world is filled with connected devices, while the digital world of cloud services and digital twins integrates them to generate massive amounts of data at scale. But data is only valuable if it can be trusted. Trusted digital interactions and transactions require knowing who and what we are interacting with. That’s why one of today’s most urgent challenges is managing the machine identities that modern digital infrastructure relies on.  

Just as employees swipe an ID badge to enter a building, machines use digital certificates and cryptographic keys to authenticate and authorize access to digital systems. These credentials function like ID badges for servers, apps, cloud services and connected devices.


Machine identities are multiplying quickly, but many organizations still rely on outdated ways to manage them. This challenge is only growing as technological advances like agentic AI and digital twinning expand the attack surface and give threat actors more ways to break in.


Learning from past weaknesses to create a stronger future


I recall a particularly costly incident. 

A few years ago, a Europe-based telecommunications provider came to DXC when its systems suddenly stopped working. Customers couldn’t make phone calls or send emails. Business operations had ground to a halt.

The DXC team quickly traced the outage to a single point of failure. For years, one employee had manually renewed the digital certificates that kept the company’s servers running. This process worked until he retired early, and no one took over the task. It went unnoticed for six months until the certificates expired and everything stopped working.



The consequences are clear: Overlooked machine identities can bring entire organizations down. Even minutes of downtime can cost millions in damage and lost revenue.




Carving a path forward to reduce risk  


The incident I described highlights a critical lesson: To protect operations and reduce risk, organizations must automate how they manage and maintain digital credentials. 

Manual processes aren’t just slow, but they’re also fragile, error-prone and increasingly unsustainable, especially with how fast machine identities are multiplying. 

In the telecom provider’s case, DXC not only restored operations quickly, but also put safeguards in place to prevent future outages. By automating the tracking and renewal of machine identities, DXC helped eliminate reliance on a single employee, ensuring business continuity and reducing risk.



According to CyberArk’s 2025 Identity Security Landscape report, machine identities now outnumber human users by more than 80 to 1 — and the gap is only expected to grow.



Automating certificate management is essential


TLS certificates that secure web servers currently expire after 398 days. By 2026, that window will shrink to 200 days; by 2027, to 100; and by 2029, to just 47.

For organizations managing thousands — even tens of thousands — of endpoints, that change means a twelve-fold increase in renewals. Manual processes can’t keep up. While shorter certificate lifespans strengthen security, they also create serious operational challenges.  



Automation not only keeps certificates current at scale, but also prepares organizations for the next wave of complexity — from the surge in AI identities to the risks quantum computers pose to today’s cryptographic algorithms.




Managing the machine identity boom from agentic AI


Managing cyber risk is becoming more complex with the rise of agentic AI.  

Each autonomous agent — whether ordering equipment for a factory manager or authorizing payments for a billing supervisor — needs its own credential to securely access APIs, databases and other systems. Manual oversight can’t keep up.

To handle the scale and speed of these interactions, organizations must adopt automated systems that can track and update machine identities in real time. This approach aligns with DXC's Zero Trust security model, built on the principle of never trust, always verify.


Without automation and proper controls, credentials are left vulnerable to cyber threats, misuse and other disruptions.



Preparing machine identities for the quantum threat


Although still in its early stages, quantum computing already poses serious risks alongside its benefits. Its speed and power could eventually break the cryptographic algorithms that secure machine identities and digital communications across enterprise networks.

To manage the risks ahead, organizations must embrace crypto agility, the ability to rapidly switch to stronger encryption when current methods become outdated or vulnerable. 

Automation is a key to crypto agility. It enables organizations to deploy new crypto algorithms and refresh keys and certificates before a new vulnerability can disrupt operations.



Overhead view of atrium with shadows on a tile floor, SAP + Databricks: Putting data and AI to work | DXC Technology Insights

 

The big picture

Protecting servers, devices and critical infrastructure is becoming harder as technology evolves and regulations shift — a challenge that will only grow with AI and other emerging technologies.

Leaders can no longer afford to manage machine identities by hand. Manual oversight leaves organizations exposed to risk and disruption, while automation keeps credentials current and systems secure as the digital landscape changes.

To stay resilient, organizations must move toward intelligent, automated systems that can keep pace with rapid growth in machine identities and ever-evolving threats.





About the author

Martin Reilly leads DXC's digital identity portfolio team to broaden and deepen our offerings globally.

Watch the replay of Martin's related LinkedIn Live session.