Build agile governance for critical infrastructure resilience

Evolving digital transformation and rising threats and attacks have led critical infrastructure to be in the middle of a fundamental shift. As power networks and grids, communications systems and transportation systems become software-defined and networked, the traditional playbook for running them has become obsolete. Asset owners and operators can no longer afford to choose between security and speed. Modern infrastructure requires governance based on the current reality.

The landscape calls upon organisations to establish clear ownership and governance aligned with secure-by-design and secure-by-default policies. Embedding security advisors alongside delivery teams ensures clear ownership and strong security governance, helping translate policy into repeatable practice and enabling faster delivery.

This shift reflects fundamental changes in how critical infrastructure operates. Power networks, telecoms networks, transport systems, water services and many other critical services are becoming software-defined, networked and automated. Technologies enabling that transformation also introduce new dependencies and attack paths. The question for regulators and operators is therefore no longer whether infrastructure should be governed or agile, but how governance can enable safe agility.

Research across technology and infrastructure markets points in that direction. Analysys Mason research on network resilience highlights how configuration errors and operational failures can produce outages affecting essential services, while its research on 5G security has warned that software-defined networking, virtualisation and the scale of connected devices create new security challenges.

At the same time, Gartner’s research on resilient and agile cybersecurity programmes explicitly connects governance with business agility, arguing that organisations need clear governance, resource prioritisation and continuity architectures to remain adaptive. Gartner has also identified regulatory volatility as a driver of more flexible cybersecurity strategies and operating models.

Prevailing threat landscape

China-linked threat campaigns Volt Typhoon and Salt Typhoon have exposed a critical governance paradox: traditional critical infrastructure oversight – built around stability, formal approvals and predictable change cycles – is fundamentally incompatible with the speed at which state-backed adversaries operate.

Data from the UK’s National Cyber Security Centre identified that over 200 CNI incidents have been reported in one year, out of which 75% were attributable to nation-state hackers, proving that such threats and attacks are no longer theoretical. Such campaigns operate on attacker timelines, instead of regulatory ones. Most often, they are searching for probing systems, exploiting credentials and vulnerabilities, while possibly maintaining persistence until conditions are suitable for major disruptions. Clearly, governance has shifted irreversibly, as organisations must choose not between security and agility, but rather redesign governance to achieve both.

One possible outcome is for organisations to adopt operationally agile governance, where regulators establish outcomes and minimum controls, while operators gain flexibility in implementation, armed with predefined risk thresholds, delegated decision-making authority, continuous monitoring and tested response playbooks. This helps invert the traditional model from ‘Did you follow the process?’ to ‘Can you prove safety and resilience remain acceptable?’

However, agility without governance creates unsafe infrastructure; while governance without agility leaves it exposed to sophisticated adversaries that operate on threat timelines. For critical infrastructure operators and regulators, architecting governance that accelerates safe decisions while containing risk may become key to defining resilience capability.

Old governance model is becoming a bottleneck

Traditional infrastructure governance was designed around relatively stable systems. A technology change could be assessed, approved, scheduled and implemented through a defined lifecycle. That model becomes harder to sustain when networks, cloud platforms, IoT devices and software-defined infrastructure change continuously.

Appledore Research’s infrastructure security research argues that increasing automation of network infrastructure is making convergence of security and network operations imperative. It points to software-defined networks, AI-driven orchestration and edge-to-cloud architectures as environments where automation creates scale and new risk.

Clearly, the implication is important, as governance cannot remain a separate administrative layer sitting above engineering. It has to become part of the operating architecture. That means moving away from governance based primarily on permission and toward governance based on risk, evidence and predefined boundaries.

Regulation must become more adaptive, not weaker

The answer is not deregulation. CNI operators still need enforceable requirements for resilience, incident response, supply chain risk, identity, access control and recovery. However, regulation specifying technical details is likely to become outdated faster due to infrastructure changes.

McKinsey’s research on national critical infrastructure resilience suggests that it is necessary to develop an organisation-wide strategy that integrates IT and OT (operational technology) security, analyse interdependencies, execute simulations and define roles in decision-making responsibilities.

A PwC report identified that stakeholders want stronger centre-led CNI coordination, with clear accountability, more robust cross-system governance and leadership that can resolve cross-departmental trade-offs. They also want a defined hierarchy of resilience objectives. Without these, government departments could continue to pursue competing objectives without aligning to a unified mission.

Thus, an agile regulatory model should establish outcomes and minimum capabilities, while allowing operators to determine how those outcomes are achieved. That could include risk-based controls rather than identical controls for every asset, continuous evidence of resilience rather than periodic compliance exercises, regulatory sandboxes for new technologies in controlled environments, pre-approved response patterns for high-risk operational scenarios, faster approval paths for changes that remain within defined risk tolerances and mandatory exercises that test whether governance actually works during disruption.

Economic scale of problem is already forcing the issue

ABI Research’s critical-infrastructure cybersecurity forecast projected global CNI cybersecurity spending at about $236 billion by 2027, representing a 14% compound annual growth rate. Its industrial research separately forecasts almost $4.6 billion in OT and IoT cybersecurity spending for industrial critical infrastructure by 2027. This investment is occurring while infrastructure operators are simultaneously adopting new technologies.

According to the IDC Asia/Pacific Security Study, 76.5% of respondents did not feel confident about their detection and response capabilities against AI-based attacks. It also noted that the risk of AI-driven risk vectors increases in verticals dealing with sensitive and confidential information, such as banking and financial services (BFSI) and healthcare, apart from critical infrastructure sectors like energy, transportation and telecommunications, where disruptions can have widespread consequences.

Waiting for certainty before approving innovation therefore creates its own risk. But deploying technology without controls creates another. Agile governance has to manage both.

Build governance into operating models

The workable model is neither a bureaucracy that slows every decision nor an innovation programme that treats controls as obstacles. It is a governance architecture embedded into operations.

Deloitte’s Critical Infrastructure Cybersecurity research indicates that IT/OT convergence is expanding cyber risks within critical infrastructures, while its cyber risk framework emphasises security, vigilance and resilience balance. A Deloitte India study indicated that more than 45% of OT cybersecurity budgets were allocated to tools while governance and culture lagged.

That imbalance matters. Buying another security tool does not resolve unclear authority, fragmented accountability or slow incident decisions.

EY similarly describes cybersecurity transformation as requiring fit-for-purpose governance, integration with business strategy and agile execution through defined sprints.

For operators, governance should be measured by operational outcomes. Engineers should be able to introduce safe changes quickly. Security teams should be able to identify whether a change increases operational risk. Executives should be able to make shutdown or isolation decisions without navigating multiple approval layers. Regulators should receive meaningful evidence without operators being forced into repetitive compliance exercises. The organisation should be able to adapt controls when technology or threats change.

From compliance to controlled adaptability

Telecoms illustrates the challenge particularly well. Kaleido Intelligence found that more than 70% of surveyed operators felt only partially prepared for future 5G threats, while forecasting mobile operator fraud and security losses would reach $45 billion in 2025 before declining as 5G security measures mature.

HardenStance’s 2026 telecoms research also highlights growing importance of resilience planning across sectors and supply chains as threats to telecom infrastructure evolve.

Lessons extend beyond telecoms

CNI governance must evolve from asking, ‘Did the operator follow the prescribed process?’ to asking, ‘Can the operator demonstrate that the infrastructure remains within an acceptable risk envelope while it changes?’ Prevailing threat landscape calls for a much more demanding standard, which coincidentally happens to be a more agile one.

The future of CNI protection will not be built by choosing governance over innovation, or innovation over governance. It will depend on designing governance systems that make responsible innovation faster, clearer and safer. Across critical infrastructure installations, agility without governance is reckless. Governance without agility is increasingly dangerous in its own right.

Anna Ribeiro Anna Ribeiro

Freelance Writer