Cisco adds AI red teaming and agent monitoring capabilities to enterprise security platform
Cisco has announced new capabilities across its AI Defense, Splunk and Cisco IQ platforms to help enterprises test AI agents for security vulnerabilities, monitor their performance and identify infrastructure risks.
The company is integrating Anthropic’s Claude Mythos into its AI Defense red-teaming framework to help security teams simulate complex, multi-step attacks against AI applications and agents before deployment. Testing will be restricted to registered and authorised targets, with operator authentication and audit trails. The capability is expected to be available to select enterprises soon.
Cisco is also expanding Splunk Agent Observability with tools to track AI agent costs against output quality. Its Tokenomics capability will use data from Amazon Bedrock and CloudWatch to link token consumption with agent performance and business outcomes. The company is also introducing evaluation models and a tuning tool that use agent activity traces and expert feedback to assess output quality and enforce guardrails. These evaluation...
Copyright of this story solely belongs to www.expresscomputer.in. To see the full text click HERE
