Amazon Web Services (AWS) recently experienced one of those incidents that make half of the internet nervous: power and connectivity issues in their Middle East infrastructure, affecting both United Arab Emirates and the neighboring region of Bahrain. The event, confirmed by the company itself, occurred after “unidentified objects” hit a facility in the UAE, causing sparks and a fire. Authorities were compelled to cut power to part of the complex while repairs and safety checks were carried out.
The incident didn’t only impact technical services. It also left its mark on daily life: financial institutions in the UAE reported temporary disruptions in apps and digital platforms, and several customers received the unwelcome message no one wants to see during working hours: back up your data and, if possible, migrate critical workloads to another region.
What happened and why did the impact extend to Bahrain?
AWS explained that the problem originated from damages in their UAE region, and that emergency response involved power cuts to contain the incident and ensure safety. Some reports indicated that the outage affected two clusters and that full recovery could take many hours, with initial estimates of around one day to restore all services to normal.
While AWS avoided publicly identifying the “objects,” technology outlets contextualized the event within a scope of regional military escalation. Over the same weekend, a surge of attacks involving drones and missiles targeted Gulf states amid rising tensions between the US, Israel, and Iran. Additional reports described damage to AWS data centers in the UAE and Bahrain linked to drones attacks, while AWS maintained a more cautious tone in their preliminary communications.
The technical consequence was the most painful aspect of cloud reliance: not just a momentary outage, but degradation of “core” services, elevated errors, and temporary limitations that force clients to activate contingency plans. The domino effect becomes clear quickly: in regions where companies and government agencies have built their digital operations on AWS, such an incident affects banks, fintechs, retailers, and public platforms immediately.
Banks and digital services: when “cloud” becomes an end-user issue
Among the most visible cases, Abu Dhabi Commercial Bank (ADCB) announced that its mobile banking and customer service centers were temporarily unavailable due to a “regional IT outage.” The bank did not directly attribute the problem to AWS, but the timing of the incident in the cloud fueled perceptions of vulnerability: when shared infrastructure experiences issues, end-users perceive it as my app has crashed, even if the root cause is hundreds of kilometers away.
Other local Emirati media also reported interruptions in banking services and digital channels during the event, reinforcing a now well-established idea: the cloud is an essential service, and its availability impacts daily operations, from paying bills to accessing loans or validating transactions.
The uncomfortable precedent: in the age of computing, data centers are also “targets”
What makes this incident especially sensitive is not just the outage itself, but the type of cause. For years, resilience discussions in the cloud focused on power failures, human errors, internal fires, or network issues. Here, the debate shifts: if the damage relates to an episode of regional violence, a disturbing door opens.
A recent analysis by the Center for Strategic and International Studies (CSIS), cited by some tech outlets, warned that in past conflicts, regional actors targeted energy infrastructure like pipelines and refineries; in the “computing era”, the risk extends to data centers, electrical infrastructure powering computing, and critical fiber points. This phrase encapsulates a shift: digital infrastructure is becoming as strategically vital as oil was decades ago.
A blow at the worst possible moment: the Gulf aims to be an AI hub
The paradox is that the incident occurred just as the UAE, Saudi Arabia, and other Gulf countries are trying to position themselves as regional AI computing powerhouses. It’s well documented how Big Tech is doubling down on investments in the region, despite geopolitical risks: Microsoft has committed over $15.2 billion between 2023 and 2029, with more than $7.3 billion already invested, much of it aimed at data centers for AI and cloud computing. Meanwhile, AWS plans multi-billion dollar investments in Saudi Arabia to develop regional cloud infrastructure, and companies like Google and Oracle are also advancing AI-related projects.
This context makes the incident doubly symbolic. On one hand, it confirms the Gulf’s desire to attract “computing” as a new strategic industry. On the other, it highlights that for an AI hub to be credible, cheap energy and land are not enough: physical security, stability, and operational resilience at levels previously taken for granted are now critical.
What companies should take away (without panicking)
For the general public, all this might seem distant… until their bank app stops working. For companies, the lesson is more direct: cloud reduces costs and accelerates deployment, but does not eliminate risk; it redistributes it.
Without delving into technicalities, three ideas often differentiate those who “survive” from those who get stuck during a regional crisis:
- Backups outside the region: local backups aren’t enough if the entire zone degrades.
- Multi-region plans for critical workloads: even in “minimal” mode, with switch-over capability.
- Real testing: a continuity plan that isn’t practiced is just a nice document.
In this case, AWS itself recommended that clients protect critical data and consider migrating workloads whenever possible—a practical way of saying: if your service can’t go down, don’t rely on a single zone without a security net.
Incident summary table
| Element | What is known (based on available info) |
|---|---|
| Location | AWS region in United Arab Emirates, also impacting Bahrain |
| Reported cause by AWS | “Unidentified objects” caused sparks and fire; power cut for safety |
| Scope | Power and connectivity outages; service degradation and phased recovery |
| Recovery time | Multiple hours; initial estimates around a day for full normalcy |
| Visible effect | Disruptions in digital services (including banking), with backup/migration recommendations |
| Context | Regional military escalation with drone and missile attacks reported by some outlets |
Frequently Asked Questions
What does it mean that AWS experiences an outage in a region like UAE or Bahrain?
It means that applications hosted there could suffer errors, slowdowns, or stop working altogether, impacting banks, businesses, and services relying on that infrastructure.
Did AWS confirm it was a military attack?
Not explicitly in their initial communications. The company mentioned “unidentified objects.” Technology reports placed the incident in a context of drone and missile attacks in the region, with information about damage related to drone strikes.
Why can a local incident affect so many services?
Because many companies concentrate critical systems in a single cloud region due to latency, regulations, or costs. If that region degrades, the impact multiplies.
What can a company do to mitigate regional incident risks?
Design continuity plans: backups outside the region, multi-region capacity for critical workloads, and regular recovery drills.