13-hour AWS outage reportedly caused by Amazon's own AI tools

Source: Engadget

Outage Overview

A recent Amazon Web Services (AWS) outage that lasted 13 hours was reportedly caused by one of its own AI tools, according to reporting by the Financial Times (link). The incident occurred in December after engineers deployed the Kiro AI coding tool to make certain changes, say four people familiar with the matter.

Kiro AI Tool

• Agentic nature – Kiro can take autonomous actions on behalf of users.
• Trigger – The bot reportedly decided it needed to “delete and recreate the environment,” which allegedly led to the lengthy outage that primarily impacted China.
• Amazon’s stance – The company says it was “a coincidence that AI tools were involved” and that “the same issue could occur with any developer tool or manual action.”
  • The outage is blamed on user error, not AI error.
  • By default, Kiro “requests authorization before taking any action.”
  • The staffer involved had “broader permissions than expected — a user access control issue, not an AI autonomy issue.”

Multiple Amazon employees told the Financial Times that this was at least the second occasion in recent months in which the company’s AI tools were at the center of a service disruption. “The outages were small but entirely foreseeable,” said one senior AWS employee.

Kiro was launched in July (announcement) and has since been pushed onto employees (inside look). Leadership set an 80 percent weekly use goal and has been closely tracking adoption rates. Amazon also sells access to the agentic tool for a monthly subscription fee.

Previous Outages

These incidents follow a more serious event from October, in which a 15‑hour AWS outage disrupted services such as Alexa, Snapchat, Fortnite and Venmo, among others (Engadget report). Amazon blamed a bug in its automation software for that outage (details).

Amazon’s Response

In response to the Financial Times report, Amazon published the following statement on its news blog (full statement):

We want to address the inaccuracies in the Financial Times’ reporting. The brief service interruption they reported was the result of user error—specifically misconfigured access controls—not AI as the story claims.

The disruption was an extremely limited event last December affecting a single service (AWS Cost Explorer) in one of our 39 geographic regions. It did not impact compute, storage, database, AI technologies, or any other of the hundreds of services we run.

The issue stemmed from a misconfigured role—the same issue that could occur with any developer tool (AI‑powered or not) or manual action. We did not receive any customer inquiries regarding the interruption.

We have implemented numerous safeguards to prevent this from happening again, including mandatory peer review for production access. While incidents involving misconfigured access controls can occur with any developer tool, we think it is important to learn from these experiences. The Financial Times’ claim that a second event impacted AWS is entirely false.

For more than two decades, Amazon has achieved high operational excellence with our Correction of Error (COE) process. We review incidents together so that we can learn from any event, irrespective of customer impact, to address issues before their potential impact grows larger.

Update

February 21 2026, 11:58 AM ET – This story has been updated to include Amazon’s full statement in response to the Financial Times report.