What preparations are necessary to accelerate the adoption of AI agents? – Visualizing and addressing "invisible risks" with Cisco.
With the rapid advancement of generative AI, we are entering an era where AI agents autonomously perform tasks. While AI agents can be a driving force for businesses, new challenges such as visibility, security, quality, and cost, exemplified below, are also becoming apparent. These include: "Shadow AI," "Agent misuse (data leakage due to unintended access, etc.)," "Operational burden proportional to the expansion of what needs to be protected, such as routes, IDs, and permissions," "Risks in the quality of AI agent responses (hallucination, etc.)," "Insufficient cost management of running applications," and "Difficulty in risk prediction due to the lack of precedents." In this seminar, we will organize the risks that companies face in the age of AI agents and introduce a comprehensive countermeasure approach utilizing Cisco's broad portfolio. Furthermore, we will provide a clear explanation of concrete usage scenarios through demonstrations of security and visibility using Cisco AI Defense and Splunk Observability. We will also introduce support from Macnica before and after implementation and share key points for safely and effectively promoting the use of AI agents.
