Amazon's AI Coding Assistant Q Compromised: Hacker Injects Data-Wiping Commands

The Breach: A Hacker's Warning Shot

In a startling development, Amazon's AI-powered coding assistant, Q, was compromised by a hacker who successfully injected potentially destructive code into the tool. The incident, which occurred in July 2025, exposed nearly one million users of the Amazon Q Developer Extension for Visual Studio Code to potential system-wiping risks 1

1

.

The hacker, using the alias 'lkmanka58', managed to submit a pull request to Amazon Q's GitHub repository, which was unexpectedly approved. The malicious commit included a prompt instructing the AI agent to "clean a system to a near-factory state and delete file-system and cloud resources" 2

2

. This compromised version (1.84.0) was unknowingly published by Amazon on July 17, making it available to the entire user base 3

3

.

Amazon's Response and Security Concerns

Upon discovering the breach on July 23, Amazon quickly investigated and released a clean version (1.85.0) the following day, removing the unapproved code 3

3

. In a security bulletin, AWS stated:

"We immediately revoked and replaced the credentials, removed the unapproved code from the codebase, and subsequently released Amazon Q Developer Extension version 1.85.0 to the marketplace." 2

2

However, Amazon's handling of the incident has drawn criticism from security experts and developers. The company initially failed to issue a public announcement and was accused of attempting to cover up the breach by quietly removing the compromised version without proper disclosure 1

1

.

Implications for AI in Software Development

This incident has raised serious questions about the integration of AI tools into software development pipelines. Corey Quinn, chief cloud economist at The Duckbill Group, commented, "This isn't 'move fast and break things,' it's 'move fast and let strangers write your roadmap.'" 4

4

The breach exposed critical flaws in how AI tools are managed and secured. It highlighted the need for robust code review processes and proper repository management practices. As Steven Vaughan-Nichols of ZDNet pointed out, the issue was not with open source itself, but rather with how Amazon implemented its open-source workflows 1

1

.

The Hacker's Motivation

Interestingly, the hacker claimed that their actions were intended as a warning rather than a malicious attack. In comments to 404 Media, they described Amazon's AI security measures as "security theater" and stated that the wiper was designed to be defective 2

2

. Their goal was reportedly to expose Amazon's inadequate safeguards and prompt improvements in security practices.

Broader Implications and User Advice

This incident serves as a wake-up call for the tech industry regarding the risks associated with AI integration in development workflows. It underscores the importance of vigilance and thorough security measures when incorporating AI tools into software development processes 4

4

.

Users of the Amazon Q Developer Extension are strongly advised to update to version 1.85.0 or later to mitigate any potential risks 5

5

. Furthermore, developers are cautioned not to unconditionally trust IDE extensions and AI assistants, highlighting the need for ongoing vigilance in the rapidly evolving landscape of AI-assisted software development.