Strategic shift in AI adoption... Compared to company-wide deployment, "business unit automation" is the key to success or failure.

The company’s AI adoption strategy is rapidly evolving.
Compared to deploying AI tools uniformly across all employees, embedding AI into specific business processes that actually encounter bottlenecks and verifying its effectiveness is considered more practical.
Recently, the case of U.S. nonprofit organization AARP and process automation platform company Appian ($APPN) exemplifies this trend well.

Appian Vice President of Product Management Jake Rank stated at the “Appian World 2026” event in Florida that, rather than broadly deploying AI roles, it is more important to use them precisely within a “controlled scope.”
He explained, “You can expect to deploy AI tools throughout the organization and hope they work, but issues centered around processes require process-centric solutions. Limiting AI to specific stages reduces the risk of misoperation and allows connecting the necessary data and task information at the right moment.”

AARP starts AI modernization with invoice approval processes

AARP is a nonprofit organization serving Americans over 50.
The organization began with high-repetition tasks like invoice approval, which require high accuracy, security, and audit tracking, launching AI-based process modernization.
AARP Platform Management Vice President Tom Cavanaugh explained that this task involves dozens of employees within the organization, so what is needed is not simple automation but a “verifiable, traceable” workflow.

The tool introduced by AARP is Appian Composer.
This tool can extract requirements from existing systems and generate operational workflows in the form of actual applications.
Its strength lies especially in allowing business users to directly view and review processes on a natural language-based interface before entering the development stage.

Cavanaugh said, “The advantage of this tool is that you can sit in front of the business leader, ask what problems they have, and demonstrate how it works on the spot.
When users see the problem firsthand and understand the possible solutions, the starting point for change is created.”

Starting from one person’s inconvenience, expanding to issues affecting over 40 people

An interesting aspect of this case is that AI-based process modernization did not stop at improving a single business.
What initially seemed like a problem faced by just one administrative assistant due to a complex invoice review process actually affected over 40 people within the organization with similar inefficiencies.

In this structure, a weekly saving of 5 hours per person can scale to the entire team, creating a cumulative efficiency gain.
Rank explained that this approach creates a “virtuous cycle” that funds the next automation project.
He said, “Each project involves assessing what technology is suitable and whether AI can solve the problem.
The result is that if each project yields a positive return on investment, these outcomes will drive the next project, forming such a cycle.”

Organizational culture change is more challenging than technology

Process modernization is both a technical and a cultural challenge.
Some argue that outdated work methods persist even if they are inefficient because the members who created them fear change.
Cavanaugh commented that the visibility provided by Appian Composer helps reduce this psychological barrier.

He explained that because business users can confirm new processes directly in natural language before coding, it promotes understanding and participation more than vague unease.
Transformations that once caused fear are now turning into opportunities for business units to proactively ask IT, “Please help us do this too.”

Rank also noted that AARP has found an effective model.
He emphasized, “The key is applying AI to specific tasks and confirming actual ROI.
Now, just review other processes in the same way to find further areas for improvement.”

AI ROI—“where to apply” is the core

This case shows that the success or failure of AI adoption depends more on the “application point” than on the technology itself.
Deploying AI across all business areas with high expectations is challenging to manage and verify.
Conversely, selecting repetitive, measurable problems and advancing automation in phases makes it easier to quantify cost reductions and productivity gains.

Ultimately, the core of AI-based process modernization is not a sweeping, comprehensive implementation, but accurately identifying organizational pain points.
The case of AARP and Appian demonstrates that when AI is rooted as a “precise business engine” rather than a “general-purpose tool,” the tangible results for companies and institutions can be significantly greater.

TP AI Notes: This article uses a language model based on TokenPost.ai for summarization.
The main content may omit details or differ from actual facts.