Cisco AI agents rollout to 90,000 employees: On-prem governance and CFO analytics

Cisco is rolling out AI agents to its approximately 90,000 employees starting in its new fiscal year, with each employee receiving a personalized assistant that handles tasks, answers questions, and routes requests to the most suitable AI model. The architecture is predominantly on-premises, emphasizing cost and data control, with dynamic model selection to optimize performance and token use. In finance, AI already assists MD&A drafting at 80%-90%, and Cisco is developing an investor-relations AI tool and a CFO cockpit dashboard to synthesize performance data and suggest actions. This deployment signals a practical blueprint for large-scale enterprise AI agent governance.

What happened

Starting in Cisco’s new fiscal year at the end of July, every employee will have access to a personalized AI assistant. The system dynamically selects the most effective AI model for each task, avoiding excessive use of frontier models. Most of the infrastructure is on-premises, enabling governance and data control through internal stacks. CFO Mark Patterson noted that the system “knows which tool is most effective and most efficient.”

Cisco is pairing the rollout with company-wide upskilling efforts and internal knowledge sharing to encourage experimentation. Patterson expects internal competition as teams discover new AI applications.

On the finance side, MD&A preparation is now 80%-90% AI-assisted. Cisco has built and is refining an AI tool for investor relations that analyzes its financial history alongside competitors’ earnings calls and can anticipate likely questions from specific analysts. Patterson and his team are also refining a “CFO cockpit,” an AI-powered dashboard that would synthesize performance data across products, geographies, and customer segments, predicting where the business is headed and recommending actions.

Why AI builders should care

Cisco’s approach shows how to scale AI agents within a large enterprise while maintaining governance and data control through on-prem infrastructure. Dynamic model routing demonstrates a practical method to balance performance, cost, and token usage when agents call multiple AI tools. The rollout includes an internal upskilling program and knowledge sharing to promote experimentation and cross-team adoption.

Plans for an investor-relations AI tool and a CFO cockpit illustrate how AI can synthesize multi-dimensional business data to guide decision-making. For AI builders, these patterns are directly applicable to enterprise deployments where data residency, cost control, and model selection are critical.

Practical implications

Organizations considering AI agent deployments can learn from Cisco’s emphasis on on-prem AI stacks to retain data control and reduce policy risk. Developing an AI-powered finance workspace (MD&A automation, investor relations tools, and dashboards) can accelerate reporting cycles and improve scenario analysis. Encouraging internal competition alongside structured upskilling can accelerate discovery of useful AI applications while managing governance.

Cisco’s broader AI infrastructure business is also growing. The company reported $2 billion in orders in fiscal year 2025 and raised its fiscal year 2026 guidance to $9 billion, reflecting growth from hyperscaler partnerships and AI-enabled networking, custom silicon, optics, and security.

Caveats

Some claims, such as exact cost allocations for the AI agents, are not separately disclosed in earnings reporting. Deployment details and governance policies are evolving as Cisco scales the rollout across the organization. Token usage estimates vary widely for agentic tasks, and Cisco’s claims reflect a specific internal deployment context. The company’s stock performance and guidance figures are based on its own projections and market conditions.

FAQs

What is Cisco's rollout of AI agents for its 90,000 employees?

Cisco is rolling out personalized AI assistants to every employee starting in its new fiscal year. The assistants handle tasks, answer questions, and route requests to the most appropriate AI model. The infrastructure is predominantly on-premises to control cost and data. Source

How does Cisco's on-premises AI architecture work?

Cisco emphasizes an on-premises infrastructure to control cost and data, with dynamic model selection that picks the best tool per use case. The system avoids burning tokens on frontier models when simpler models suffice. Source

What is the CFO cockpit dashboard and how does it help finance?

The CFO cockpit is an AI-powered dashboard intended to synthesize performance data across products, geographies, and customer segments. It predicts where the business is headed and recommends actions, helping finance teams make faster, data-driven decisions. Source

What is Cisco's investor relations AI tool?

Cisco is building an AI tool that analyzes its financial history alongside competitors’ earnings calls. It anticipates likely questions from specific analysts, helping the investor relations team prepare more effectively. Source