Am I an AI Power User? [2026 Survey]

Am I an AI power user? When AISA asked 95 people starting the AI fluency assessment "What best describes your AI journey so far?", 33.7% said AI is already part of how they work, and another 13.7% said they'd be lost without it. That means 47.4% place themselves in the two most integrated stages. But the remaining 52.6% are still experimenting, finding a rhythm, or haven't started at all. Where you fall on that spectrum tells you more than any label.

1. Haven't started20.0%19
2. Experimenting15.8%15
3. Finding my rhythm16.8%16
4. It's part of how I work33.7%32
5. I'd be lost without it13.7%13

95 people answered, 2 August to 6 August 2026. Source: AISA research, aisa.to

Am I an AI Power User?

The honest answer is that "AI power user" has no universal definition, so the question is better reframed: how deeply is AI woven into the way you actually get things done? AISA's research gives us five self-reported stages that map the journey from zero to full integration. If you picked "It's part of how I work" or "I'd be lost without it," you're operating in the territory most people mean when they say power user. If you picked anything else, you're not behind; you're at a specific, identifiable point on a path that others have already walked.

The five stages people chose, from earliest to most integrated:

1. Haven't started (20.0%)
2. Experimenting (15.8%)
3. Finding my rhythm (16.8%)
4. It's part of how I work (33.7%)
5. I'd be lost without it (13.7%)

The largest single group, at 33.7%, has already made AI part of their working pattern. That's the stage where AI stops being a novelty and starts being infrastructure. The smallest group, at 13.7%, has gone further: they describe a dependency, a workflow so shaped by AI that removing it would leave a gap.

What "power user" actually signals

Being an AI power user isn't about hours logged in ChatGPT. It's about whether you can decompose a task, pick the right tool, verify the output, and fold the result into your actual work. The people in the top two stages have likely built those habits, whether or not they'd use the phrase "power user" themselves.

The middle is where most growth happens

The 16.8% who said "Finding my rhythm" are in the most active learning phase. They've moved past random experiments and are starting to notice which prompts work, which tools fit, and where AI saves real time. This is the stage where AI fluency starts to compound.

How Many People Say AI Is Part of How They Work?

One in three people AISA asked, 33.7%, chose "It's part of how I work," making it the most common answer by a wide margin. This stage describes someone who doesn't open an AI tool out of curiosity; they open it because it's the next step in a process they've already designed.

What "part of how I work" looks like in practice

People at this stage tend to have repeatable patterns. They might draft with AI, review with their own judgement, and iterate using prompt chaining or structured output formats. The key marker is consistency: they don't use AI on Monday and forget about it on Tuesday. It's embedded.

The training gap behind the number

This 33.7% figure sits against a backdrop where formal support is thin. According to the Microsoft and LinkedIn Work Trend Index (May 2024), only 39% of people who use AI at work have had AI training from their company. BCG's AI at Work 2025 report (June 2025), covering over 10,600 workers in 11 countries, found that only 36% of employees feel adequately trained in AI use. Many of the people who say AI is part of how they work got there on their own.

That matters because self-taught adoption is uneven. Some people build strong habits; others build fragile ones. Without a way to measure where you actually stand, "it's part of how I work" could mean anything from sophisticated multi-tool workflows to copying and pasting ChatGPT output without checking it. This is exactly why measuring AI fluency matters more than measuring adoption.

How Many Would Be Lost Without AI?

13.7% of the 95 people AISA asked said they'd be lost without AI. That's 13 people who have built their workflows so thoroughly around AI tools that removing them would create a real disruption.

Dependency vs. fluency

It's worth separating two things here. Dependency on AI isn't automatically a sign of expertise. Someone could be lost without AI because they've never learned to do the underlying work without it, or because they've built such efficient AI-augmented processes that going back would be genuinely slower and worse. The difference between those two situations is critical thinking: can you evaluate what the AI gives you, or do you accept it by default?

The concept of cognitive surrender describes the risk at the extreme end. When someone stops questioning AI output entirely, their dependency becomes a liability rather than a strength. A genuine AI power user is dependent on AI by choice, not by default, and can articulate why each part of their workflow uses AI rather than a manual alternative.

What this means for teams

If you manage a team, the 13.7% figure is a signal worth paying attention to. These are the people who will feel the most friction if tools change, licenses get pulled, or policies shift. They're also likely your most productive AI users. Understanding who they are and how they work lets you protect what's working and spread those patterns to others. Posts like How Has AI Changed Your Job? explore this shift in more detail.

What Does Each Stage of the AI Journey Look Like?

Each of the five stages represents a distinct relationship with AI tools, not just a frequency of use but a depth of integration. Here's what patterns tend to characterise each one.

Haven't started (20.0%)

One in five people AISA asked said they haven't started using AI. This isn't a small fringe. It's a fifth of the group. These are people who may be aware of AI tools but haven't found a reason, an opportunity, or the confidence to begin. They're not necessarily resistant; they may simply lack a clear entry point.

The Microsoft and LinkedIn Work Trend Index (May 2024) found that 66% of leaders say they wouldn't hire someone without AI skills. That stat doesn't mean people who haven't started are unemployable. It means the expectation is shifting, and the window to start is still open but narrowing.

Experimenting (15.8%)

Experimenters are in the "try everything" phase. They might ask ChatGPT a question one day, generate an image the next, and try a coding assistant the day after. The hallmark of this stage is breadth without depth. Nothing has stuck yet as a habit.

The risk here is stalling. Experimentation is fun, but without a deliberate shift toward repeatable use, it stays recreational. The guide on prompting skills is a practical starting point for turning random experiments into structured practice.

Finding my rhythm (16.8%)

This is the transition stage. People here have found one or two use cases that work and are starting to build consistency. They might draft emails with AI every morning or use it to summarise meeting notes every afternoon. The pattern isn't fully formed, but it's emerging.

What separates this stage from the next is reliability. People finding their rhythm still have days where they forget to use AI or fall back to old methods. The shift to "part of how I work" happens when the AI step becomes the default, not the exception.

It's part of how I work (33.7%)

The largest group. These people have integrated AI into their workflows to the point where it's no longer a separate activity. They don't "use AI" as a distinct task; they do their work, and AI is part of how that work gets done. This is the stage where prompt engineering stops being a skill you practise and starts being a skill you apply.

I'd be lost without it (13.7%)

The most integrated stage. These people have rebuilt their workflows around AI to the degree that removing it would require significant restructuring. They're likely using AI across multiple tasks, multiple tools, and multiple parts of their day. The question for this group isn't whether they use AI well; it's whether they can articulate and transfer what they do.

How Do You Move From Experimenting to a Rhythm?

The shift from experimenting to finding a rhythm comes from narrowing your focus and repeating what works. Experimenters try many things; people with a rhythm have committed to a few.

Pick one workflow, not one tool

The most common mistake in the experimenting phase is tool-hopping. A better approach is to pick one workflow you do repeatedly, such as writing status updates, reviewing documents, or researching a topic, and use AI for that same workflow every time for two weeks. The tool matters less than the repetition.

Build a prompt you reuse

People who move from experimenting to rhythm almost always have at least one prompt they've refined over multiple uses. It might be a system prompt for a specific task, a template they paste in, or a conversation starter they've memorised. The act of refining a prompt forces you to think about what you actually want from AI, which is the core of AI fluency.

Verify before you trust

The other shift that marks the transition is learning to check AI output. Experimenters often accept or reject output wholesale. People with a rhythm have developed a sense for what to trust, what to verify, and what to rewrite. The guide on critical thinking with AI covers this in detail.

Measure where you stand

If you're asking "am I an AI power user?" the most direct answer comes from measuring your skills across the dimensions that matter: prompting, critical thinking, technical understanding, workflow integration, and responsible use. The AI fluency assessment gives you a score across all five, with evidence-linked feedback on where you're strong and where to focus next.

Is It a Problem if You Have Not Started?

No, but the window is getting smaller. 20.0% of the people AISA asked said they haven't started, which means one in five people taking an AI assessment are doing so precisely because they know they need to begin.

The pressure is real but manageable

The hiring signal is clear: 66% of leaders say they wouldn't hire someone without AI skills, according to the Microsoft and LinkedIn Work Trend Index (May 2024). That doesn't mean you need to become an expert overnight. It means starting matters more than starting perfectly.

Starting is easier than it looks

The gap between "haven't started" and "experimenting" is the smallest gap on the entire journey. It requires one action: open a tool, type a question, see what comes back. The gap between "experimenting" and "finding my rhythm" is harder because it requires discipline. But the first step is genuinely low-effort.

If you're in the "haven't started" group and want a structured way to begin, the AI Fluency Academy guides walk through each skill area from scratch. And if you want to know where you stand before you start building skills, the AI fluency assessment takes about 15 minutes and gives you a detailed breakdown.

For teams dealing with uneven adoption, AI Change Management: The People Side covers how to support people at every stage, and AI Skills Gap Statistics puts the broader gap in context.

About this research

This finding comes from AISA's own research. AISA asks a short research question of people as they start the AISA AI fluency assessment, one question at a time. The question in the chart above was asked from 2 August to 6 August 2026, worded exactly as shown and with the answers in the order shown, and 95 people answered it. Each person counts once: when someone answered more than once, their latest answer is the one counted.

More findings from this research are collected in AISA research and data.

How to cite this research: AISA (2026). "Am I an AI Power User? [2026 Survey]". AISA research. https://aisa.to/blog/am-i-an-ai-power-user

---

Related reading: What Does AI Fluent Mean? [Defined], the definition behind the score. Related reading: How Has AI Changed Your Job? [2026 Data], what people say AI has actually shifted in their work. Related reading: Is Prompt Engineering Still Relevant? [2026], whether the skill still matters as tools evolve.

Frequently Asked Questions

What makes someone an AI power user?

An AI power user has integrated AI into their regular workflows to the point where it's a default part of how they work, not an occasional experiment. In AISA's research, 33.7% of people said AI is part of how they work and 13.7% said they'd be lost without it. The distinguishing factor is consistency and depth of integration, not just frequency of use.

How do I become an AI power user?

Start by picking one repeatable task and using AI for it consistently. Refine your prompts over time, learn to verify AI output rather than accepting it wholesale, and gradually expand to more tasks. Measuring your current AI fluency with an AI fluency assessment gives you a concrete starting point and shows you which skills to build next.

Is it bad if I have not started using AI?

It's not bad, but it's worth starting soon. 20.0% of the people AISA asked said they haven't started, so you're not alone. The gap between "haven't started" and "experimenting" is the smallest step on the journey: it takes one task and one tool. The pressure is increasing, with 66% of leaders saying they wouldn't hire someone without AI skills according to the Microsoft and LinkedIn Work Trend Index (May 2024), but starting now still puts you ahead of the curve.