EmergePH · Original Research · 2026 Edition

The AI Behavior Benchmark

How Filipino Teams and Leaders Make AI Stick

Almost every organization tracks whether people are using AI. Almost none track the behavior that decides whether it actually holds. This benchmark measures that behavior directly, across leaders and teams, and shows where the gap between belief and reality really sits.

Takes about 8 minutes. Your individual answers are never shown to anyone at your organization.

Illustrative
The behavior gap
What leadership believes
What teams actually do
14710
What It Measures

Five behaviors decide whether AI adoption holds

Not tool usage. Not sentiment. The benchmark scores the actual behaviors that separate organizations where AI sticks from organizations where it stalls, read through how Filipino teams and leaders actually work under pressure.

01

Ownership

Spotting a work problem and using AI to solve it without being told to.

02

Modeling

Leaders making their own AI use visible, including what did not work.

03

Experimentation

Trying a new way of working with AI and treating a miss as information, not failure.

04

Learning Autonomy

Teaching yourself the next AI skill instead of waiting to be trained.

05

Value Articulation

Describing your work in terms of the value AI helped create, not just the tasks it touched.

The conditions underneath

Psychological safety

Admitting you used AI, or that you do not know how, carries no cost.

Job safety belief

People believe using AI to do more makes them more valuable, not more replaceable.

A clear leadership signal

Leadership has said, specifically, what the AI investment means for people's jobs and futures.

Who It's For

Built for leaders and teams, answering the same questions

There are two ways in. Both lead to the same benchmark.

On your own

Take part independently

Come to the benchmark on your own, as a leader or as part of a team, and contribute your read to the published research.

Through your organization

Take part through a link

If someone at your organization shared this with you, your answers are grouped with theirs so your organization can see its own gap, once enough people have answered.

Leaders and teams answer the same underlying questions. Neither layer's view is treated as more true than the other's. The distance between them is the finding.

If You Lead an Organization

See the distance between what you believe and what your people report

You cannot measure this from your own seat. That is the whole reason it goes unmeasured.

1

You take the benchmark. Eight minutes. At the end you get a private link for your organization.

2

You send that link to your team. Four or more of them answer within seven days. With your own response, that makes a group of five, which is the minimum any finding is ever reported from.

3

You get an org gap readout. What you believe is happening, set beside what your team reports, behavior by behavior. Free, and private to you.

Individual answers are never shown to anyone, including you. You see grouped results only. If fewer than four of your team respond, no readout is produced and nothing is shown, which is what makes it safe for them to answer honestly.

What You Get

A read on your own behavior, and a hand in real research

The reading is a thank-you, not the point of taking part.

Nascent → Embedded A band, not a number: Nascent · Developing · Practicing · Embedded

At the end, you get a personal AI Behavior Readiness reading, given as a band with a short behavioral profile, not a hard score. What lands where is part of what makes it worth taking.

The real value is the published research itself. It gives you a market-wide reference point for a number your own organization has never had, and one no one in the Philippines has published before.

Your individual answers are never shown to anyone at your organization. Only grouped results are ever shared, and only once enough people have answered. Nothing is ever reported from a group smaller than five.

How Your Data Is Used

Grouped, anonymized, never traced back to you

Your responses are grouped with others' and anonymized before anything is reported. Published findings never identify a person or a small group. Your work email is used only to send you what you specifically opted into, nothing else. This handling follows the Philippine Data Privacy Act of 2012.

Read the full privacy statement for the complete detail on what is collected, why, and how long it is kept.

Who's Behind This

EmergePH, and the method behind the number

Almost every organization we walk into can tell us how many people have an AI licence. Almost none can tell us whether behavior changed. We built this instrument because we kept needing a number that did not exist, and because the Philippine market has been asked to take a great deal on faith about AI adoption without anyone measuring what is actually happening inside the work. We are publishing what we find, including the parts that are inconvenient for people who sell AI programs. We sell one of those programs.

Eli Harrell
Founder, EmergePH

Who we are

EmergePH is a Philippines-based behavior change consultancy. This benchmark is our own research, published in the open.

Method

Behavioral self-report across leadership and team layers, scored on a shared 1 to 10 scale.

Not Ready for Eight Minutes

Get the findings when they publish

You do not have to take part to read what comes out of it.

The benchmark publishes as an open report. If you would rather just read it, we send it out through Lead Human, our letter on what leaders are actually doing about AI, behavior first. No pitch, and you can leave whenever you like.

Send me the findings

Take Part

Add your read to the research

Eight minutes, one read on your own AI behavior, and a contribution to research that names the gap most organizations cannot yet see.

Take the Benchmark
© 2026 EmergePH. All rights reserved.