AI Tools Triggered December Amazon Service Disruption, Report Says

In an era where leading corporations worldwide are integrating AI across their operations, occasional hiccups are inevitable.

Such a situation reportedly occurred with Amazon last December, as detailed by the Financial Times. The disruption affected Amazon Web Services (AWS), a critical component of the internet infrastructure, lasting for 13 hours. An Amazon spokesperson later described this incident as a minor interruption impacting “one of our two Regions in Mainland China.” However, sources indicated to the Financial Times that the disruption arose from engineers permitting the Kiro AI system to execute certain tasks, which ultimately led the AI to “delete and recreate the environment.”

It’s important to note that this incident was not comparable in scale to the significant AWS outage that occurred last October.

Additionally, AWS published a blog post to clarify perceived inaccuracies in the Financial Times’ coverage. An AWS representative also informed Reuters that the event was brief and attributed to “user error,” not AI malfunction. This view suggests that if the Financial Times’ account holds true, the focus of accountability lies with the engineers who permitted the AI’s actions rather than the AI itself. The spokesperson emphasized that the December episode did not affect essential services as the October outage did and confirmed that no customer complaints arose from this issue.

An Amazon spokesperson shared the following statement with Mashable:

“This was an extremely limited event last year when a single service (AWS Cost Explorer — which helps customers visualize, understand, and manage AWS costs and usage over time) in one of our two Regions in Mainland China was affected for 13 hours. This event did not impact compute, storage, database, AI technologies, or any other of the hundreds of services that we run. We are also not aware of any related customer inquiries resulting from this isolated interruption. In both instances referenced, the root cause was user error — specifically misconfigured access controls — not AI error. Kiro puts developers in control — users need to configure which actions Kiro can take, and by default, Kiro requests authorization before taking any action. Following the December incident, AWS implemented numerous safeguards, including mandatory peer review for production access, enhanced training on AI-assisted troubleshooting, and resource protection measures.”

Recent months have seen a rise in high-profile outages on the internet. Most notably, YouTube recently experienced a global outage. Other notable outages include incidents involving Verizon, Cloudflare, Microsoft 365, Google Cloud Platform, Microsoft Azure, and TikTok.

Experts offer differing opinions on whether internet outages are becoming more prevalent. However, one thing is evident: as websites and applications become increasingly dependent on a small number of cloud providers like Amazon Web Services, individual outages can trigger far-reaching, cascading effects throughout the internet.

UPDATE: Feb. 20, 2026, 8:24 p.m. EST The story has been updated with an additional statement and denial from Amazon Web Services. Based on the AWS statement, further contextual information has been provided regarding the disrupted region.

UPDATE: Feb. 20, 2026, 12:36 p.m. EST This article has been revised to clarify that Amazon has attributed the outages to human error, not AI.