How to read this list.

AI change management is not a single category of software. It is a stack: a framework for thinking about the change, a diagnostic for finding what is actually broken, and a set of platforms for acting on it. The mistake we see most often is buying a platform before the framework or the diagnostic. The shape of the stack is well documented in Gartner's AI research, which tracks frameworks, diagnostics, and adoption software as separate categories rather than one.

Each entry below names what the tool is, who it is best for, and where it tends to fail when used in isolation. Kotter's eight-step model, included below, was first published in Harvard Business Review's 1995 article "Leading Change" and remains one of the most-cited change frameworks in HBR's catalog.

01 - AI-first change framework

Average Robot 3P diagnostic.

Best for: Mid-market marketing, creative, and talent teams trying to align AI spend with business value.
What it is: A diagnostic across people, process, and platform that names where AI adoption is breaking and what to do next. Built for the AI-at-work problem, not retrofitted from a generic change model.
Watch-outs: Not a SaaS platform. The output is a written read and a path forward, not a dashboard.

02 - Classic change framework

Prosci ADKAR.

Best for: Structured human-side change inside large organizations.
What it is: Five-stage model (awareness, desire, knowledge, ability, reinforcement) for moving individuals through a change. The most widely adopted change management certification in the world.
Watch-outs: Pre-dates generative AI. Strong on the human transition, light on AI-native failure modes like prompt hygiene, model selection, and shadow tooling.

03 - Classic change framework

Kotter's 8-Step Process.

Best for: Leadership-driven transformation programs that need a clear narrative arc.
What it is: Eight steps from creating urgency through anchoring change in the culture. Useful for framing AI change as a sustained leadership program rather than a tool rollout.
Watch-outs: High-altitude. Needs to be paired with operational tools that act on each step.

04 - Organizational diagnostic

McKinsey 7-S.

Best for: Diagnosing alignment across strategy, structure, systems, shared values, style, staff, and skills.
What it is: A diagnostic that surfaces misalignment between the soft elements of an organization (style, staff, skills, shared values) and the hard ones (strategy, structure, systems). Useful for stress-testing whether an AI strategy is supported by the rest of the org.
Watch-outs: Diagnostic only. Does not prescribe the AI-specific moves to close the gaps.

05 - Adoption software

WalkMe / Whatfix / Pendo (digital adoption platforms).

Best for: Driving in-app adoption of a specific AI tool or workflow after the rollout has shipped.
What it is: In-app guidance, walkthroughs, and adoption analytics overlaid on existing software. Tells you who is using what, where they drop off, and where to inject training.
Watch-outs: Measures activity, not value. Strong adoption metrics on an AI tool that is solving the wrong problem still ends in failure.

06 - Enablement software

ChangeGear / Whatfix Mirror (enablement and training).

Best for: Scaling AI literacy and tool-specific training across thousands of employees.
What it is: Content, sandbox environments, and tracking for AI upskilling at scale. Most useful inside organizations that have already picked an AI tool stack.
Watch-outs: Solves the training gap, not the strategy gap. Buy after diagnosis, not before.

07 - Vendor-native adoption

Microsoft 365 Copilot Adoption Tools.

Best for: Organizations standardizing on Microsoft Copilot across knowledge work.
What it is: Adoption scorecards, usage analytics, and playbooks from Microsoft for getting Copilot used inside Outlook, Word, Excel, and Teams.
Watch-outs: Vendor-aligned. Tells you how to drive Copilot usage, not whether Copilot is the right tool for the job.

08 - Vendor-native adoption

Google Workspace AI / Gemini Enterprise rollout kits.

Best for: Teams standardizing on Google Workspace and Gemini.
What it is: Adoption playbooks, training paths, and admin telemetry for rolling Gemini out across Workspace.
Watch-outs: Same caveat as Microsoft - solves the rollout, not the strategy.

09 - Learning and upskilling

Bridge / Degreed / Cornerstone (learning platforms).

Best for: Building AI fluency as a long-term capability inside the org.
What it is: Curated learning paths, skill tracking, and certification across AI literacy, prompt craft, and role-specific upskilling.
Watch-outs: Slow to land if not paired with concrete work. Training without a workflow to apply it to fades fast.

10 - Hybrid

Custom AI-first playbooks (in-house).

Best for: Organizations with the maturity to build their own AI operating model.
What it is: Internally authored playbooks combining a framework (Prosci, 3P), a governance model, and a tool stack tailored to the org. Often the end state for companies that started with off-the-shelf tools.
Watch-outs: Heavy lift to author and maintain. Needs senior internal owners or it drifts.

How to sequence the stack.

Diagnose first. Use an AI-first diagnostic (3P, or a tailored McKinsey 7-S read) to find where AI is producing value and where it is producing cost.
Pick a framework. Prosci ADKAR for the human transition, Kotter for the leadership arc. Both can sit on top of the diagnostic.
Buy a platform only after the diagnosis points to one. Adoption analytics, enablement, or learning platforms all assume you already know which AI workflows are worth driving.
Measure value, not activity. Tool dashboards optimize for usage. Pair them with a value metric tied to business strategy, AI strategy, or customer strategy.
Re-diagnose on a cadence. AI at work shifts every quarter. A static change program built in Q1 is already partially wrong by Q3.

For a self-serve diagnostic across the same dimensions, start with the free AI Alignment Snapshot. The longer argument sits in our book, The Elephant in the Algorithm.

Questions people ask.

What are AI change management tools?

They are the software, frameworks, and diagnostic methods leaders use to move an organization from AI swirl to strategic intent. Some are platforms (adoption analytics, enablement, learning). Some are frameworks (Prosci ADKAR, Kotter's 8-step, McKinsey 7-S). Some are diagnostics built specifically for AI at work, like Average Robot's 3P (people, process, platform).

Is Prosci ADKAR still the right framework for AI change?

ADKAR is still useful for the human side of any change (awareness, desire, knowledge, ability, reinforcement). It was not built for AI specifically, so it does not name AI-native failure modes like model drift, prompt hygiene, or shadow tool sprawl. Most teams pair ADKAR with an AI-first diagnostic that surfaces those gaps.

Do I need a change management platform or a change management partner?

A platform measures adoption and pushes enablement. A partner finds why adoption is stuck and what to do about it. Buying a platform without a diagnosis is the most common reason AI rollouts stall - the dashboard reports low usage and no one knows why.

What is the difference between AI change management and digital transformation?

Digital transformation tends to be a multi-year program with clear deliverables (new CRM, new e-commerce stack). AI change management is continuous: the technology keeps shifting, the workforce keeps re-skilling, and the org has to rewire how value is measured. The frameworks for one do not automatically work for the other.

Where should a mid-market team start?

Start with a diagnosis, not a tool purchase. Map where AI activity is happening, where it is producing value, and where it is producing cost. That tells you which framework to apply and which platform - if any - is worth buying. Average Robot's AI Alignment Snapshot is a free version of this.

AI change management tools and frameworks.