Cisco, the global leader in networking, security, and infrastructure, announced a suite of AI-powered observability tools at the Splunk .conf25 event in Boston, United States.
This includes a unified data fabric, a machine data lake, and a time series foundational model for anomaly and root cause analysis.
Together, the companies aim to help enterprises turn machine-generated data into actionable AI-powered intelligence.
Powered by Splunk, the data platform the company acquired last year, Cisco announced a new AI-powered Splunk observability agent. It deploys AI agents across the entire incident response lifecycle while monitoring both its performance and quality.
For instance, the AI troubleshooting agent is offered as part of Splunk’s Observability Cloud and AppDynamics platform, to autonomously analyse incidents and identify the root cause of issues, enabling teams to resolve them quickly.
Other agents help teams set up alert correlation, summaries of alerts, trends, impacts, and more to troubleshoot faster.
“With the latest innovations in Splunk Observability, we are empowering enterprises to proactively monitor their critical applications and digital services with ease, resolve issues before they escalate, and ensure the value and outcomes they derive from observability are commensurate with the cost,” said Patrick Lin, senior vice president and general manager of Splunk Observability.
Observability is the ability to measure a system’s internal health and performance by analysing metrics, logs, and traces. It applies in various fields, including IT operations, software development, business operations, and others.
The networking and security firm also announced observability tools for these AI agents themselves. This enables continuous monitoring of their quality, security, health, and costs, ensuring they perform as intended.
The company announced that most of these features will seamlessly integrate with several solutions and platforms offered by them and Splunk.
New Cisco Data Fabric, Time Series Foundation Model
Cisco also announced Cisco Data Fabric, powered by the Splunk platform, which helps enterprises efficiently handle machine data at scale.
Enterprises can utilise the Data Fabric for various AI use cases, such as training AI models or correlating multiple streams of both machine data and business data to generate insights.
Machine-generated data is the continuous streams of information produced by sensors, servers, applications, and network devices. This includes network packet counts, CPU/GPU utilisation, readings from data centre sensors, query response times, error codes, and more.
The data flows continuously, often generating terabytes per day from a single enterprise environment, and captures the real-time health and performance of digital infrastructure.
Cisco stated that the framework enables data management across the edge, on-premises, and cloud environments, including SecOps, ITOps, DevOps, and NetOps. It can also federate data across sources like Amazon S3, Apache Iceberg, Delta Lake, Snowflake, and Microsoft Azure, with support for additional sources next year.
Besides, the company also announced a new Machine Data Lake in Splunk, which provides an ‘AI-ready foundation’ for analytics and training AI models.
The company released a Time Series Foundation Model, pre-trained to perform pattern analysis and reasoning across machine data in the time series format. The model allows for anomaly detection, forecasting, and automated root cause analysis across the Cisco Data Fabric.
“Every company has massive volumes of this machine data, but it’s been largely left out of AI for a few reasons: LLMs don’t speak the language of machine data, the information is spread across disparate silos, and the expertise and costs involved can be prohibitive. As a result, we’ve only begun to scratch the surface of what we can do with AI,” said Jeetu Patel, chief product officer at Cisco.
Patel mentioned that although the industry has been successful in training AI models using human-created data such as text, it has not yet been fully explored with data generated by machines, which motivated the recent announcements.
The Time Series Foundation Model will be made available on Hugging Face, the open-source AI model repository, starting in November 2025.
Splunk was acquired by Cisco in 2024 for $28 billion. Their competitors in the AI-enabled observability space include companies such as New Relic. Dynatrace, Datadog, Elastic, and others.
[The writer is in Boston attending Splunk .conf25 at the invitation of Cisco. Editorial coverage remains independent.]
