Microsoft Edge team automates quality checks as vibe-coded extensions flood in
Microsoft's Edge team has become the latest group inside the company to acknowledge it is struggling to keep up with a flood of AI-generated code, and has responded by automating its quality assessments. The team said the increase in so-called vibe-coded browser extensions has outpaced its ability to review submissions manually, according to The Register.
The admission follows similar acknowledgments from other Microsoft teams that have found themselves overwhelmed by the volume of code produced with AI assistance. The pattern points to a broader problem inside the company: tools that make it easier to write and ship software are also making it harder to vet that software before it reaches users.
Vibe coding refers to the practice of describing a feature in natural language and letting an AI model generate the implementation, often with the developer reviewing only loosely, if at all. Applied to browser extensions, the approach lets people produce add-ons quickly, but the resulting code can carry quality problems that a human author might have caught earlier in the process.
The Edge team's response has been to automate quality assessment rather than try to hire or train its way out of the backlog. The details of that automated pipeline, including what checks it runs and how it triages the results, have not been fully disclosed.
The Register's report is based on the team's own statements, and the full text of the article was not accessible for this summary. The specific numbers behind the increase in extensions, the failure rates the team is seeing, and the timeline for the automated review system have not been confirmed.
What remains unclear is whether automated quality checks can keep pace with a generation method that is itself automated, and whether other browser vendors facing the same flood will follow Microsoft's lead. Microsoft has not said whether the approach will be extended beyond Edge extensions to other AI-heavy codebases.
Microsoft's Edge team automating quality checks signals that AI-generated code volume is overwhelming even major software vendors' manual review processes.