The Economics of Scale
Starting at the end of July 2026, Cisco is deploying personalized AI agents to its entire 90,000-person workforce. This is not a pilot program; it is a standardized infrastructure shift intended to change how the company produces work. To manage the massive volume of requests from 90,000 users, Chief Product Officer Jeetu Patel has implemented a model routing mechanism. This system directs complex, high-stakes tasks to expensive frontier models while offloading routine work to smaller, more efficient, and less costly alternatives.
While Cisco has not disclosed the total cost of this deployment, secondary sources estimate the annual AI token bill at approximately $900 million. This figure, which should be viewed as an unofficial estimate rather than a corporate disclosure, breaks down to roughly $200 per employee per week. To maintain control over these costs and ensure data security, Cisco has opted to run much of this model routing infrastructure on-premises rather than relying exclusively on third-party cloud providers, as noted in a Fortune interview with CFO Mark Patterson.
The Workforce Tradeoff
This technological integration is occurring alongside a significant shift in human capital. In May 2026, Cisco announced approximately 4,000 job cuts, representing less than 5% of its workforce. CEO Chuck Robbins framed these reductions as a strategic realignment toward silicon, optics, security, and AI in a company blog post. Patterson has explicitly stated that these cuts were not savings-driven, but rather a reflection of changing operational priorities.
The practical impact is already visible in high-level administrative tasks. Patterson noted that 80% to 90% of the first draft of the Management’s Discussion and Analysis (MD&A) section in Cisco’s public SEC filings is now produced by AI. This shift is being mirrored in the development of a “CFO cockpit,” an AI-powered dashboard that synthesizes performance data across products and geographies to forecast business trajectories and recommend specific actions.
Why Cisco is the Template
Cisco is increasingly viewed as the enterprise template for AI adoption because it is balancing aggressive infrastructure investment with clear revenue targets. The company raised its FY2026 AI revenue target to $4 billion, with a minimum of $6 billion projected for FY2027. This financial confidence is supported by a massive increase in AI infrastructure orders, which rose from $2 billion in FY2025 to $9 billion in guidance for FY2026, according to Cisco’s Q3 FY2026 earnings report. Investors have responded: CSCO stock is up 52% year-to-date as of July 2026, outperforming its sector’s 14.7% gain.
By building the infrastructure to support agents internally and tying the rollout to specific, measurable revenue goals, Cisco provides a blueprint other large organizations are already studying. The question is not whether other enterprises will follow this template, but how quickly.
What Remains Unknown
Several variables remain. The actual cost of the 90,000-employee deployment is undisclosed, and the $900 million annual token estimate — sourced from secondary reporting on remarks by Jeetu Patel, not from Cisco’s own filings — serves only as a proxy for scale. Whether the model routing approach can hold its cost advantage as usage grows across 90,000 employees is untested at this magnitude. And while Patterson describes the 4,000 job cuts as unrelated to savings, the timing of workforce reduction alongside full-scale AI deployment is the kind of pattern that invites scrutiny from the people whose jobs are changing.