Kyndryl deploys AI system handling 10 million IT risks annually, targets 90% reduction in enterprise outages

Enterprise technology services provider Kyndryl has launched a new agentic artificial intelligence capability aimed at helping large enterprises detect and prevent technology failures before they escalate into business disruptions, underscoring the growing corporate push toward predictive IT operations and automated infrastructure management.

The new patented capability, integrated into the company’s Kyndryl Bridge platform, is already being used by more than 1,400 enterprise customers globally. Kyndryl said the platform generates over 16 million AI-led operational insights every month and has demonstrated up to a 50% reduction in IT incidents, while helping customers collectively save nearly $3 billion annually through avoided outages and lower maintenance expenditure.

Rising cost of enterprise downtime

The rollout comes as enterprises worldwide face mounting financial losses from unplanned outages, cyber incidents and increasingly complex hybrid-cloud environments. Industry studies estimate that downtime in mission-critical systems can cost large corporations millions of dollars annually in lost productivity, supply-chain disruption and reputational damage.

Against this backdrop, enterprises are accelerating investments in observability platforms, predictive analytics and AI-led automation tools capable of identifying infrastructure stress before failures occur.

Kyndryl said its latest feature uses AI agent-assisted root-cause analysis to continuously monitor operational signals across applications, infrastructure and enterprise networks. The system has already been deployed across more than 200,000 customer devices globally.

AI agents move IT operations from reactive to predictive

“By embedding AI agents in Kyndryl Bridge for proactive risk detection, we are fundamentally shifting enterprise IT operations from reactive firefighting to predictive prevention,” said Xerxes Cooper, Global Leader, Kyndryl Delivery.

“In large enterprise environments, outages are rarely caused by a single event. They emerge from millions of interconnected signals across applications, networks and infrastructure. Our platform is designed to correlate those signals in real time, identify the underlying risks early and help technology teams intervene before operations are affected. That reduces downtime, lowers operational costs and improves resilience across increasingly complex hybrid environments,” Cooper added.

According to the company, the platform is capable of handling early-stage detection for more than 10 million incidents annually and, in certain deployments, has demonstrated up to a 90% reduction in mission-critical production outages.

Faster root-cause analysis

Kyndryl also said the platform significantly cuts the time required for root-cause analysis of major technology failures. Investigations that traditionally took weeks can now be completed within hours through AI-assisted analysis and automated correlation of operational data.

The company added that while AI systems generate recommendations and predictive insights, Kyndryl experts continue to validate findings before implementation in customer environments, particularly in complex multi-vendor enterprise setups.