A 13-hour outage in amazon web services in december was reportedly triggered by the company's kiro ai coding tool, according to sources familiar with the matter. Amazon attributes the incident to user error rather than ai malfunction. This marks at least the second such disruption linked to amazon's ai tools in recent months.
In mid-december, amazon web services (AWS) experienced a 13-hour interruption to a system that allows customers to explore service costs, primarily affecting parts of mainland china. Four people familiar with the matter told the financial times that engineers had deployed the kiro ai coding tool to implement changes, and the agentic tool, which can act autonomously, decided to "delete and recreate the environment," leading to the outage.
Amazon disputed the characterization, stating it was a "coincidence that ai tools were involved" and that "the same issue could occur with any developer tool or manual action." The company emphasized that the incident stemmed from "user error, not ai error," noting the engineer had "broader permissions than expected—a user access control issue, not an ai autonomy issue." By default, kiro requests authorization before actions, but no second approval was sought, contrary to usual procedure.
Multiple amazon employees reported this as at least the second recent incident involving ai tools. One senior AWS employee told the financial times, "We’ve already seen at least two production outages [in the past few months]. The engineers let the ai [agent] resolve an issue without intervention. The outages were small but entirely foreseeable."
The second incident involved amazon's earlier amazon q developer tool and did not impact customer-facing services. AWS launched kiro in july to advance beyond basic code generation, setting an 80 percent weekly usage goal for developers and tracking adoption closely. The company sells access to kiro via monthly subscription and reports strong customer growth.
These events follow a larger 15-hour AWS outage in october 2025, caused by a bug in automation software, which disrupted services including alexa, snapchat, fortnite, venmo, and openai’s chatgpt. AWS, which generates 60 percent of amazon’s operating profits, has since implemented safeguards like mandatory peer review and staff training after the december incident.
