Generational

Media

Cyber resilience is infrastructure, and it needs an owner

Power, payments, telecom and health share software and vendors, so cyber resilience needs an owner and operators should act before regulators require it.

Cyber resilience in critical sectors should be treated like roads or the grid: shared infrastructure that someone is accountable for. Today it is treated as the private concern of each operator, with oversight split among agencies and regulators by function. That works when risks are separate. It fails when power, payments, telecom and health depend on the same software and the same few vendors, because then one fault becomes many outages. We write here only about defending civilian services and keeping them running.

Shared dependencies make shared risk

A bank, a hospital group and a power distribution company have little in common, yet they buy from overlapping sets of suppliers: operating systems, database software, cloud services, managed service providers and the firms that maintain their industrial control systems. A flaw in one widely used product, or a compromised update from one vendor, reaches all of them at the same moment. Risk that looks independent on each balance sheet is correlated in practice.

Spending also protects unevenly. A large operator can afford a security team, regular testing and a second site. A district hospital, a municipal utility or a small payments processor often cannot, and yet it may sit on the path that a bigger system relies on. Each organisation’s spending protects itself and does little for the network around it. That is the usual signature of an infrastructure problem, and it is why a purely market answer falls short.

Who holds which piece today

India already has the beginnings of an institutional answer. CERT-In, the national computer emergency response team under the Information Technology Act, 2000, receives incident reports, issues alerts and coordinates response. Its directions of 2022 require specified kinds of incident to be reported within hours, not weeks. NCIIPC, the National Critical Information Infrastructure Protection Centre, works under the National Technical Research Organisation and is the nodal agency for systems whose failure would damage national security, the economy or public safety. Sector regulators, such as the Reserve Bank of India for banks and payments, add their own technology and security expectations.

The Digital Personal Data Protection Act, 2023 adds a different lever. It places duties on any organisation that decides how personal data is used, including taking reasonable security safeguards and notifying the Data Protection Board and affected individuals after a personal data breach. A hospital or lender that holds records on millions of people therefore has a legal reason, not only a commercial one, to invest in security.

Responsibility is divided by function, but failure travels by dependency, and nobody is yet asked who owns the dependency.

That is our central argument. The agencies above each hold a piece. What is thin is ownership of the question that matters in a crisis: if this shared vendor, product or provider fails on Tuesday, which sectors stop, in what order, and who coordinates the response? Until that question has an answer, resilience depends on goodwill and on whichever operator happened to ask it first.

What private operators should do before being told

Waiting for a mandate is a poor strategy, because mandates arrive after incidents and are written in a hurry. An operator that wants to be ahead of the rule can do the following now.

  • Keep a live inventory of software, suppliers and service providers, and mark the ones that would stop operations if they failed.
  • Test recovery, not only backup. Restore a critical system from an offline copy, time it, and fix what breaks.
  • Run drills that include vendors, and where possible the sector regulator, so that contact lists and decisions are tested before they are needed.
  • Write vendor contracts that require prompt notice of incidents and access to security information about the product.
  • Separate operational technology, such as plant controls, from office networks, and review the few places where they connect.
  • Report incidents early and share technical indicators with peers and CERT-In, even when the rule would allow more time.

Supply chain trust deserves more weight than it gets. Buyers should ask suppliers for a software bill of materials, a list of the components inside a product, so that when a flaw in a common component is announced they can tell within hours whether they are affected. In telecom, India has already taken the view that equipment should come from trusted sources, which treats vendor origin as a security question. Other sectors will face the same logic.

One position here may draw disagreement: mandatory reporting alone will produce minimal reporting. Firms that fear penalties and press coverage disclose late and thinly. Rules that pair the duty to report with confidentiality, no-blame review of honest mistakes and fast, useful help in return will get better information sooner.

What we are watching is whether sector drills reach the small operators, the district hospital and the municipal water or power utility, and whether anyone has a plan to pay for their basic protection. The most serious weakness in a shared system is usually its smallest connected member.

More from Media