After a Run of Tech Outages, CISA Says Resilience Matters Whatever the Cause
A run of high-profile technology disruptions is a reminder that systems need to be built to keep working under stress, according to Nick Andersen, acting director of the US Cybersecurity and Infrastructure Security Agency (CISA). Speaking at the FedTalks event, he described the last couple of days as highly disruptive, pointing to flight ground stops, a global Microsoft Outlook outage and banking slowdowns.
Andersen said the events illustrate themes CISA has been raising for about a year. Organisations should be able to demonstrate resilience against disruption whether it stems from an IT outage, a bad patch, a modernisation update gone wrong or a malicious cyber incident. In his view, the cause matters less than the outcome. People simply want clean water to keep flowing, groceries to stay on shelves and daily life to continue without constant, significant disruption.
He also noted that online discussion is currently dominated by the so-called "vulnpocalypse", the worry that artificial intelligence is speeding up the discovery of software vulnerabilities. As a result, he suggested, ideas such as building and testing for resilience are perhaps being forgotten. He said this is a point of emphasis in CISA's "CISA 2030" strategy.
On the Federal Aviation Administration ground stop last week, Andersen said CISA had found nothing to indicate malicious cyber activity. He referred questions about the National Security Agency's internal communications system going down to the Defense Department. Separately, the Senate Homeland Security and Governmental Affairs Committee is due to hold a hearing on Michael LaSapio, nominated as CISA's permanent director.
Why it matters
Many businesses build their security plans around attackers, yet the outages Andersen cited, including one where no malicious activity was found, show that ordinary failures can halt operations just as effectively. Anyone relying on a single cloud email provider, bank or booking system is exposed, and the decision this forces is to define what must keep working and test it, not just prevent intrusions. The attention on AI-driven vulnerability discovery risks crowding out that resilience work.
What to do
- Identify the systems your business cannot operate without, such as email, payments and booking tools.
- Test your fallback process for each critical system, including how staff communicate if email is down.
- Stage and test patches and updates before wide rollout, since a bad patch can cause an outage.