AI Toolkit

This resource is for the MillerKnoll dealer network.

Theory & POV

A model for how AI actually works. In practice, not in theory.


Before tactics, you need a model. This page explains the formula behind every decision in this toolkit, the flywheel that builds collective capability, and why the most dangerous AI risk isn't hallucination — it's assuming AI adoption is automatic.

The Formula

Why fast alone doesn't work.


A calculator doesn't make you better at math. Better at math makes you better with a calculator. AI works the same way. The tool multiplies whatever capability and readiness you bring to it.

C × R = Value
C
Capability
What AI can do with your inputs
×
multiplied by
not added to
R
Readiness
Your skill, context, and judgment
=
Value
Real business output
Not just activity

What happens at R = 0

If your organizational readiness is zero (no practice, no shared language, no feedback loops), the AI capability multiplier produces zero. High capability × zero readiness = zero value. That's not a technology problem. It's a readiness problem.

What this means in practice

Every investment in this toolkit is an investment in R. The prompts, the habits, the manager rhythms: they're all ways to raise your readiness so that when AI capability grows (and it will), you're able to multiply it into something real.

Historical Context

MillerKnoll has been here before.


Every era of design brought new tools that seemed to change everything. The job was always the same: understand people, design around them. The pattern is worth studying.

1950
Problem

Eames Shell Chair

Fiberglass manufacturing unlocked mass production of organic forms. The constraint wasn't vision. It was materials. Charles and Ray Eames didn't ask "how do we make chairs faster?" They asked "what shapes can we now make that we couldn't before?"

New tool → new forms previously impossible

1994
Problem

Aeron Chair

Ergonomic research revealed that knowledge workers were developing back problems from sitting eight-plus hours a day. The Aeron didn't make sitting faster. It redefined what sitting well meant and built a category in the process.

New understanding → new category

2025
Now

Gemma Recliner

Real-time biometric data adapts to your body as your needs change throughout the day. The sensor data doesn't replace the human body's signals. It amplifies them into better support. That's what AI is supposed to do for thinking.

Amplification of human signal → better outcomes

"The question isn't whether AI changes the work. It's whether you're the one deciding how."
The Flywheel Model

How teams go from individual tactics to collective capability.


A single person using AI well is a tactic. A team where everyone uses AI well and shares what they learn is a system. The Flywheel is the model for getting from one to the other.

1

Individual Tactic

Someone on the team tries a prompt. It works. They save it. That's the seed: a single moment of earned insight.

2

Shared Learning

That prompt gets shared in a team standup or Slack channel. Two more people try it. One of them improves it. The learning compounds.

3

Collective Practice

The team builds a library. They develop shared language. New hires get onboarded into a system, not just handed a subscription.

4

Operational Literacy

AI thinking becomes part of how the team operates. It shows up in client meetings, proposals, retrospectives. Not just individual tasks.

5

Community of Practice

The flywheel turns on its own. People share what works. The organization's capability raises the floor for everyone. This is Org B.

Common Questions

The skeptic's FAQ.

If you've heard any of these questions in a team meeting, here's how to answer them.

The Eames Shell chair was dismissed as a novelty in 1950. The ergonomics movement was dismissed as overhead in the early 1990s. Neither passed. The pattern in design history is consistent: tools that fundamentally change what's possible don't pass. They become the baseline. The question isn't whether AI becomes baseline. It's whether your team gets there before or after your competition.
Tell the truth, and frame it accurately. "We used AI as a thinking partner to develop this analysis. Every recommendation was reviewed and refined by our team." That's honest, and it's actually a differentiator. It signals that you're investing in your process. The clients who push back on AI use are increasingly rare; the ones who ask why you're not using it are becoming the norm.
You verify it. This isn't different from any other research or analysis. You don't publish a client proposal without reviewing the numbers. AI output should be treated as a strong first draft from a knowledgeable but fallible colleague: useful, worth reading carefully, not appropriate to pass along unreviewed. Build verification into your workflow the same way you would with any junior contributor.
The nuanced answer: AI will change which parts of jobs are valuable. Work that required an hour of research can now be done in minutes. That hour is now available for the higher-order thinking that AI still can't do: building trust, reading a room, making judgment calls in ambiguous situations. The people who get displaced are typically those who refused to adapt. The people who thrive are those who figured out how to multiply their expertise with the new tools.
You understand the model

Now let's use it.

The Toolkit has prompts, tactics, and use cases built for the MillerKnoll dealer context. Start with the Stakeholder Intelligence Prompt. Highest immediate impact.

Go to the Toolkit → Manager's Guide