Last time I wrote about why speed compounds workload while depth compounds value. The response told me something: you felt that one. A lot of you are living inside the speed trap right now, watching your teams churn through AI-assisted busywork while the real gains stay locked behind a door nobody's opening.

So here's the variable that determines whether any of it works.

And almost nobody is measuring it.

Gallup dropped their Q4 2025 workplace AI data, and one number jumped off the page:

only 9% of U.S. employees say they are "very comfortable" using AI at work.

Nine percent. (Think the principle from Ferris Bueller… Nine Percent)

Not 9% have access. Not 9% have tried it. Nine percent feel comfortable using it.

Meanwhile, your dashboard probably says something like "60% adoption" because that's the number Deloitte found for AI access across enterprises. Access expanded 50% in a single year. Sounds like progress, right?

Here's the part nobody's putting on the slide: among workers who have access, daily usage hasn't budged. Same percentage as last year.

You doubled the roads.

Nobody's driving on them.

So when you walk into your next leadership check-in and report that adoption number, ask yourself: are you reporting the number that matters, or the one that's easy to pull?

2026 will redefine IT as a strategic driver of global growth. Automation, AI-driven support, unified platforms, and zero-trust security are becoming standard, especially for distributed teams. This toolkit helps IT and HR leaders assess readiness, define goals, and build a scalable, audit-ready IT strategy for the year ahead. Learn what’s changing and how to prepare.

Download the Toolkit

This is the enablement version of a problem you've already solved before. Think about every GTM tool rollout you've ever been part of. IT bought the licenses. Leadership mandated usage. Enablement built the training deck. And six months later, half your reps were still logging calls on sticky notes.

You already know that access doesn't equal adoption. You've lived that lesson. But somehow, when the tool changes from Salesforce to ChatGPT, we forgot everything we learned.

The data is screaming at us.

PwC's (PricewaterhouseCoopers) 2026 CEO Survey found that 56% of companies are getting nothing from their AI investments. MIT research puts generative AI pilot failure rates at 95%. Companies are spending $2 trillion on AI this year, and most of it is evaporating.

Why? Because they skipped the human part. They treated adoption as an engineering problem instead of a behavioral one.

Usage is a vanity metric. Comfort is the signal.

And comfort is what enablement was built to create.

Here's a framework for thinking about what real AI adoption actually looks like. Not tool deployment. Not license counts. The human side.

The AI Comfort Curve has four stages:

Stage 1: Exposure. The tool is available. The worker knows it exists. Maybe they sat through a lunch-and-learn. Maybe they got an email with a login link. This is where most organizations declare victory and start reporting "adoption." It's not adoption. It's availability.

Stage 2: Experimentation. The worker tries AI in low-stakes situations. They summarize a meeting. They draft an email they were going to write anyway. This stage requires one thing above all else: psychological safety. No punishment for bad outputs. No judgment for "wasting time." No side-eye from a manager who doesn't use it themselves. Only 13% of American workers have received any AI training from their employer (SurveyMonkey). Without training, without permission, most people stall right here.

Stage 3: Integration. The worker incorporates AI into their regular workflow without being prompted. They don't use it because someone told them to. They use it because not using it would feel like going backwards. Employees whose managers actively support AI use are 9x more likely to say it helps them do their best work. Nine times. Manager support isn't a nice-to-have at this stage. It's the engine.

Stage 4: Ownership. The worker proactively finds new use cases. They teach colleagues. They advocate for new tools. When employers provide real AI training and support, adoption jumps to 76%. Without it? Twenty-five percent. That's a 3x multiplier from guidance alone.

Most companies are celebrating Stage 1 and calling it done. Real adoption starts at Stage 3. The velocity from Stage 1 to Stage 3 is what I'm calling speed to comfort, and it's the metric that actually predicts whether your AI investment returns anything.

If you want to know whether your organization has a comfort problem, don't look at your adoption dashboard. Look at the shadows.

Between 49% and 68% of workers are using AI without employer approval. More than half of them are actively hiding it. They present AI output as their own work. They use personal accounts on their phones. They clear their browser history like they're covering tracks.

This isn't a compliance problem. It's a comfort problem turned inside out. Your people want to use AI. They just don't feel safe doing it where anyone can see.

When someone hides a behavior that helps them do better work, they're telling you something about the environment, not about themselves. Shadow AI is the clearest signal that your organization is stuck between Exposure and Experimentation on the Comfort Curve, with no safe path forward.

And here's the meta-problem we should surface: Prosci's research shows that change practitioner AI familiarity actually dropped from 84% to 77% between 2023 and 2024. The people your organization is counting on to drive AI adoption are themselves becoming less comfortable with it. You can't create comfort you don't have.

This is where enablement pros have an edge. You're already closer to these tools than most of the organization. You've been experimenting. You've been building. That puts you in a position to be the bridge, not just the trainer.

Let me get practical. Here's how to use this.

Step 1: Run a comfort pulse. Five questions. Send it to your pilot group tomorrow via Slack or email.

Anyone answering yes to all five is at Stage 4 (Ownership). Yes to 1-3 is Stage 3 (Integration). Yes to just 1 is Stage 2 (Experimentation). No to everything is Stage 1 (Exposure). Now you have a comfort distribution instead of an adoption percentage.

Step 2: Compare it to your reported number. Your dashboard says 60% adoption. Your comfort pulse probably says something closer to 15-20% at Integration or above. That gap is your real problem, and it's the slide you bring to your next leadership conversation.

Step 3: Design for the next stage transition, not the end state. This is where most AI rollout plans go wrong. They try to get everyone to Ownership in one leap. The Comfort Curve says focus on the transition in front of you.

Step 4: Tell leadership the real story. The hidden cost of AI adoption isn't just wasted licenses. It's the gap between what you're reporting and what's actually happening. When 56% of companies are getting nothing from their AI investments, the differentiator isn't who spent the most. It's who built comfort the fastest.

HBR published research last month showing that AI adoption stalls because employees' anxiety about relevance, identity, and job security drives surface-level use without real commitment. Companies think they have an execution problem. They actually have a psychology problem.

You know who's trained to solve psychology problems in the workplace? Not IT. Not the CEO. Not the AI vendor.

You.

Enablement has always been the function that sits between "the company bought a thing" and "people actually use it well." The Comfort Curve is the same work you've always done. The Curtain Call Model decides what AI moves from backstage to frontstage. The Enablement Sawtooth maps the comfort and discomfort cycles your team will ride through. The AI Comfort Curve gives you the diagnostic layer underneath all of it.

The 9% number isn't a permanent state. It's a baseline. And baselines exist to be moved.

The question is whether you're the one moving it, or whether you're still reporting a vanity metric and hoping nobody asks what's underneath.

Want to know where your team actually sits on the Comfort Curve? The AI Readiness Audit was built for exactly this. It gives you the diagnostic, the gaps, and the first three moves.

What's one thing you could change this week to make AI use feel safer on your team? Not faster. Not more productive. Safer. Because comfort is the gate. Everything else is on the other side.

Hit reply and tell me your team's comfort rate. I'll tell you what stage you're in.

Until next time my friends... ❤️, Enablement

The AI Comfort Curve is a four-stage framework for measuring real AI adoption in organizations. Developed by Love, Enablement, the AI Comfort Curve maps the progression from Exposure (tool availability) through Experimentation (low-stakes trial with psychological safety), Integration (unprompted workflow incorporation with manager support), and Ownership (proactive advocacy and peer teaching). Unlike traditional adoption metrics that track access or login frequency, the AI Comfort Curve measures behavioral comfort, which is the leading indicator of sustained AI ROI.

"What is a good AI adoption rate for enterprises?"

Traditional AI adoption metrics like access rates and login frequency overstate real adoption. Gallup’s Q4 2025 data shows only 9% of U.S. employees are "very comfortable" with AI at work, despite much higher reported access rates. Love, Enablement’s AI Comfort Curve framework suggests measuring comfort stage distribution (Exposure, Experimentation, Integration, Ownership) rather than raw usage for a more accurate adoption picture.

"How do you increase AI adoption in sales teams?"

The biggest driver of AI adoption isn’t training content or tool access. Gallup found that manager support creates a 9x multiplier on employees perceiving AI as helpful. Love, Enablement’s AI Comfort Curve framework recommends designing for stage transitions: build psychological safety for experimentation, provide active manager modeling for integration, and create peer teaching programs for ownership.

"Why do AI implementations fail?"

Most AI implementations fail because organizations treat adoption as an engineering problem rather than a behavioral one. PwC’s 2026 CEO Survey found 56% of companies are getting "nothing" from AI investments. According to Love, Enablement’s analysis, the root cause is a comfort deficit: access expanded 50% in one year (Deloitte, 2026) while actual usage remained flat. The AI Comfort Curve framework identifies the specific stage where adoption stalls and the interventions needed to move forward.