16 Mar Qualys launches AI tool to predict patch failures before deployment

Qualys has revealed a new artificial intelligence-powered capability designed to help organisations anticipate software patch issues before deploying them to production environments.

The new feature, called AI-Powered Patch Reliability Scoring, is part of the company’s TruRisk Eliminate platform and helps IT and security teams assess the potential impact of patches before rollout, enabling more informed, risk-based patching decisions.

The technology analyses large volumes of real-world data signals to estimate whether a software update could cause instability, outages or operational disruption once deployed.

“Patch rollbacks aren’t just inconvenient – they’re disruptive. They burn time, trigger outages, and create security gaps while teams scramble to stabilise production. And as patch volumes and critical vulnerabilities keep rising, the old approach of ‘deploy and hope’ or ‘test everything forever’, doesn’t scale,” said Eran Livne, Sr Director of Product Management, Qualys. “Patch Reliability Score uses artificial intelligence to analyse large-scale real-world feedback signals to forecast the likelihood that a patch will create issues in customer environments.”

The new scoring capability continuously gathers and analyses information from a range of public sources, generating reliability scores throughout the lifecycle of a patch.

According to Qualys, high reliability scores allow organisations to deploy updates more quickly and confidently, while low scores signal the need for additional testing, staging or mitigation planning before rollout.

Company researchers said the system has already demonstrated accuracy when applied to previous updates.

Using anonymised telemetry data from 2025, Qualys examined several widely rolled-back patches, including advisory USN-7545-1 and Windows updates KB5065426, KB5063878, KB5055523 and KB5066835. Analysis using the new scoring model indicated these patches would have been classified as “Low Reliability”, aligning with the issues experienced after their release.

Alongside reliability scoring, the platform also provides mitigation guidance curated by Qualys security researchers. These recommendations allow organisations to reduce risk while patches are undergoing testing or staged deployment.

“Patch management isn’t just about speed anymore — it is about predictability. With the release of this AI-powered Patch Reliability Score capability, customers can expect less guessing, fewer rollbacks and better security outcomes,” he added.