Cisco Incorporates AI Assistant into Their Cloud App Security Service

Cisco has enhanced its Panoptica cloud-native application development software with the addition of a generative AI Assistant, as well as improved security measures.

In an effort to enable customers to manage cloud-native application security issues more efficiently and promptly, Cisco is introducing an AI assistant based on generative algorithms.

As explained by Vijoy Pandey, Cisco’s Outshift advanced development group’s senior vice president, in a blog post , the Cisco AI Assistant for Panoptica is designed to assist companies in streamlining, prioritizing, and rectifying risks and misconfigurations within multi-cloud environments.

The Panoptica package by Cisco, a product of Outshift and available since September, serves as a cloud-native security platform. This platform is specially intended to protect cloud applications from the development phase to the deployment phase , with a particular emphasis on safeguarding containerized, microservice applications running on platforms like Kubernetes.

The platform empowers customers to create and enforce security protocols using tools such as Terraform, and it scrutinizes application activity to identify and prevent threats instantly, mentions Cisco. It includes features found in intrusion detection and prevention systems, tailored specifically for cloud-native settings.

The newly integrated AI Assistant comprehends common, everyday language and provides immediate, custom help in prioritizing, investigating, and rectifying a customer’s specific security problems, says Pandey. “Hence, you can make requests like ‘What are my critical vulnerabilities?’ and ‘Assist me in comprehending this attack path and how to repair it.’ As a result, it possesses cognizance and intelligence about your live environment that includes all the data Panoptica monitors regarding your posture, vulnerabilities, and attack paths,” Pandey penned.

“What’s more thrilling is that this Generative AI-based User Interface has a ‘lens-like’ feature that enables you to receive answers limited to whatever section of the Panoptica User Interface that you concentrate the ‘lens’ on, including detailed Attack Paths, thus eliminating the burden of restating that context yourself,” Pandey noted.

Moreover, it accomplishes all this with an architecture that adheres to the principles of responsible AI, according to Pandey. “For instance, the Cisco AI Assistant for Panoptica does not permit your Panoptica data to be stored by any large language model,” he composed.

Outshift has enhanced Panoptica with a new security and protection suite named GenAI Protection, in addition to its AI Assistant. The main objective of GenAI Protection is to provide enterprises with robust protection for their generative AI-based applications against the major security threats and attack methods aimed at LLMs. The identification of these major risks is based on comprehensive security frameworks such as Open Worldwide Application Security Project Top Ten, MITRE, and NIST, as described by Pandey.

Pandey goes on to explain that the GenAI Protection suite lends greater insight into training and fine-tuning data, models, and certain vulnerable artifacts such as AI Bill of Materials (AI BOM). It also aids in preventing issues related to model inference like prompt inject and insecure prompt handling. In essence, it works to identify, block, and report suspicious prompts made by users in Panoptica under API security.

Leave a Reply

Your email address will not be published. Required fields are marked *

Previous Article

Amazon Cloud Chief Compares Generative AI Hype to the Dotcom Bubble Era

Next Article

How AI Tools Like GitHub Copilot are Transforming the Way Coders Think

Related Posts