Most companies are still early.
Much of the global conversation around AI remains fixated on model breakthroughs and timelines to AGI. Despite the headlines, most organizations are still relatively early in their adoption journey.
We set out to share our unique experience as a large global technology company that has transformed its business into an AI-first organization — with over 40,000 employees across a wholly-owned global portfolio, who have together built more than 60,000 AI agents.
We believe this places us among the largest operators of AI agents globally whose core business is not the sale of AI services to other companies.
The classic power law holds.
Approximately 2% of active AI agents drive a disproportionate share of business impact. The implication is clear — the first priority for every organization is to double down on that 2%.
Organizations should also use data-driven analysis to identify — and then nurture — the agents that are trending toward the power-law group. Without this signal, high-potential agents quietly stay underused while attention scatters across the long tail.
The same 20 use cases, everywhere.
Across industries, geographies and languages — with no central mandate — portfolio companies independently converged on the same set of high-ROI agentic use cases.
While not a comprehensive list, these use cases are a good starting point that all organizations should implement. They also offer a window into the future: tasks that will likely become “default AI” for every business within the next few years. Below are three examples of these use cases, and the full list can be found in the report.
Agents map to seniority levels.
AI agent complexity falls into four tiers that map closely to human employee seniority levels.
Senior-level agents have the largest total number of users. When looking at daily usage, however, it splits almost evenly between senior and junior agents — revealing that simpler AI agents carry significant weight in daily tasks.
Departmental breakdown.
We identified 54 distinct AI agent tasks across corporate functions. A full breakdown by department and task is included in the downloadable report.
Data analytics and market intelligence claimed the largest share at 18%, where AI's ability to surface business trends from large internal and external data sets is extremely valuable. Operations follows at 15%, for tasks such as forecasting supply or managing inventory. Notably, the third largest share of AI agents — 14% — sits outside any formal department, representing employees' personal AI assistants.
Three bands of hours saved.
Productivity agents — where ROI is measured in hours saved — divide cleanly into three distinct bands.
The majority — 82% — deliver modest but meaningful gains: under 20 hours saved per month, typically personal AI assistants for individual employees. The middle tier (17%) saves between 20 and 173 hours monthly — roughly up to one FTE — by automating tasks and redirecting human effort toward higher-value work. At the top, less than 1% operate at a different scale entirely, delivering the equivalent of thousands of hours of monthly work.
A small group of outliers.
Value agents — measurable in revenue growth or cost reduction — follow the same shape. A handful deliver outsized gains.
Under $1M in annual value
For example, reducing annual audit costs.
$1M–$10M in annual value
For example, a customer-facing assistant that answers niche vacation rental questions.
Tens of millions of dollars annually
One portfolio company used AI agents to manage communications and onboarding for a new third-party affiliate marketplace. Projected annual revenue for this new marketplace is $83M.
Most models are now good enough.
Across-the-board improvements mean nearly every modern model can handle nearly every agent task. The challenge is no longer capability — it's cost discipline.
Our internal agentic AI platform, Toqan, offers ten different AI models to our 40,000+ employees. We observe that advancements in AI mean that the latest cutting-edge models are only needed for the most complex or difficult AI agent tasks.
We have also found that users are reluctant to switch models once their agent works. As a result, introducing new models creates a trade-off: employees may use unnecessarily expensive, state-of-the-art models for tasks that could easily be handled by cheaper alternatives.
AI costs are volatile and hard to predict. We've found it's more effective to focus on optimization and empower individual business lines to make their own cost decisions. Our solution is a two-tier approach:
A three-step adoption framework.
We have developed a three-step process for driving adoption of agentic AI at large organizations. The need for such a framework is real, as most large organizations have struggled to move from experimentation to scale.
In Part Two of the report, we share a detailed breakdown of what needs to happen at each step of the process.
From AI colleagues to AI-led organizations.
After scaling to 60,000 AI agents, we now have a clear view of what an AI-driven organization looks like in practice — and what comes next.
What we observe across our portfolio is transformation rather than total disruption. We see three trends:
This report focuses on implementing AI within existing ways of working — but that is just the starting point. The deeper transformation lies ahead: the rise of autonomous, AI-enabled organizations, where core business functions like sales, customer support, and operations are largely executed and coordinated by networks of AI systems, with minimal human oversight.
Comparisons are often made to how the invention of electricity impacted factories. Simply substituting steam motors or water wheels with electric motors in existing factory designs offered only relatively smaller gains. Proponents argue that it is only when companies start re-building core functions from the ground up with AI in mind that businesses will experience the biggest gains from AI.