You hear that AI can save you hours of work, so you decide to give it a try. You choose a task that normally takes an hour, explain what you need and enter a prompt. The result is underwhelming.
You refine your instructions and try again. The next result is better, but still not what you need. Twenty minutes later, you find yourself thinking, I could have done this myself by now.
So, you do. The next time the same task comes up, you skip AI and return to the way you have always done it.
I have been hearing versions of this story more frequently. People are experimenting with AI, but when it does not immediately make them faster or better at their jobs, they become frustrated and retreat to familiar ways of working.
The problem is that they may be drawing the wrong conclusion. They have not necessarily discovered that AI cannot help them. They may simply have encountered the learning curve.
Using AI is Not the Same as Being AI Literate
AI adoption has happened remarkably quickly, but adoption and proficiency are two different things.
A 2026 AI literacy study from AISA offers an interesting snapshot of that gap. Among 1,017 people who completed its assessment, nearly half fell into the category AISA defines as “AI Literate.” They were using AI but still had significant gaps in understanding how it works, where it fails and how to use it safely. Forty-four percent scored below a functional level in their understanding of AI fundamentals.
That distinction is becoming increasingly important professionally. LinkedIn’s labor-market research reports that U.S. job postings requiring AI literacy skills grew 70% year over year, extending well beyond technology roles.
You do not have to become an AI engineer. But knowing how to open an AI tool and enter a prompt may no longer be enough. The more important skill is learning how to use these tools thoughtfully, evaluate what they give you and determine where they can genuinely improve your work.
We May Be in the Messiest Part of the Learning Curve
One reason people may be giving up too quickly is that we tend to judge new technology based on whether it makes our lives easier right away.
History suggests that is not always how transformative technologies work.
In an episode of The Daily examining AI’s potential impact on jobs, New York Times chief economics correspondent Ben Casselman discusses the “J-curve,” a concept economists use to explain what can happen when a major new technology enters the workplace. Productivity may initially decline as companies and workers figure out how to use it. Eventually, people become more proficient, workflows change and productivity begins to rise.
Think about the early internet. The ways we now use it for communication, shopping, banking and work did not appear fully formed the moment people went online. Individuals and organizations had to experiment and determine where the technology was actually useful.
AI may be putting us through a similar period of experimentation, only much faster.
If your first attempts to incorporate AI into your work feel inefficient, that does not necessarily mean you are doing something wrong. Learning something new often slows us down before it speeds us up.
The question is whether you are willing to remain in that uncomfortable stage long enough to get better.
Before You Give Up, Get Curious About Why
When I work with someone who is resisting a change, I want to understand what is underneath that resistance. The same principle applies to AI.
If you are avoiding it, what are you telling yourself?
- I don’t have time to figure this out.
- I’m going to get it wrong.
- My boss expects me to know more than I do.
- The way I’ve always done this works just fine.
Or perhaps the fear goes deeper: If I help my company use AI more effectively, am I helping it figure out how to do my job without me?
There is enormous uncertainty about how AI will affect jobs, so those concerns are understandable. But it is important to pay attention to what your self-talk causes you to do.
If you tell yourself you are bad at AI, you may avoid using it. If you decide learning it is pointless because everything changes too quickly, you may stop experimenting. If you assume a disappointing result proves AI has nothing to offer you, you return to the comfortable way of doing things.
Your beliefs start driving your behavior. And avoidance makes it difficult to develop a capability that is becoming increasingly valuable in the workplace.
Stop Trying to “Learn AI”
Part of the problem may be the goal itself.
“Learn AI” is overwhelming. The technology is changing too quickly and encompasses too much for most professionals to master comprehensively. Instead, make the goal smaller.
Choose one task you already understand well and perform regularly. It might be analyzing information, preparing for a meeting, developing an outline, summarizing research, reviewing data or creating a first draft. Then experiment.
Because you understand the work, you are qualified to judge the result. If the first output is poor, resist the urge to conclude that the tool does not work. Ask why it fell short.
- Did you provide enough context?
- Were your instructions specific enough?
- Would examples help?
- Are you asking AI to accomplish too much in one step?
- Could you break the task into smaller pieces?
- Or is this simply a task where AI does not add much value?
Refine your approach and try again. The objective is not to make AI perform every part of your job. It is to identify where it can make you better at your job.
Once you find one application that consistently creates value, move to another. Fluency develops through repeated use, evaluation and adjustment.
Continuous Learning Is Becoming a Career Skill
For much of our careers, professional learning followed a fairly predictable model. You developed expertise, periodically added new skills and relied heavily on the knowledge you accumulated through experience. But that model is changing.
Continuous learning has become a career survival skill. The skills required for our jobs are changing too quickly for learning to remain something we do only when we attend a conference, earn a certification or take a course.
AI makes that especially visible. The tool you learn today may change next month. A capability that did not exist six months ago may suddenly become relevant to your work.
The objective cannot be to reach a point where you are finished learning. It is to become increasingly comfortable learning while the environment around you continues to change.
We Don’t Know How Much Time We Have to Adapt
There is another lesson from the rise of the internet worth considering.
In The Daily discussion, Casselman points out that while the internet created enormous opportunity, it also eliminated or dramatically reduced certain types of work. Typing pools disappeared. Travel agents became less common. Banking, retail and countless other industries changed.
Yet much of that disruption happened gradually and was spread across the economy. Many workers had time to see changes coming, develop new skills, change directions or move into emerging areas of opportunity.
We do not yet know whether AI will give us the same amount of time.
AI is evolving extraordinarily quickly, and we are still early in this structural shift. We do not know which jobs will change most, which new jobs will emerge or how quickly it will happen.
That uncertainty can be a reason to start learning now. We have previously explored what AI hesitancy reveals about how leaders respond to uncertainty and change. But engaging with AI is only the beginning. The next challenge is staying engaged when learning it becomes frustrating.
Stay in the Learning Curve
Nobody knows exactly what an AI-enabled workplace will look like five years from now. You do not need that answer to take action today.
You can become more curious about what AI can and cannot do. You can notice when fear or frustration is causing you to retreat. You can experiment with one part of your work rather than trying to master an entire technology. And you can develop the habit of continuously learning as the tools evolve.
AI literacy is not a destination you suddenly reach. It is a capability you build.
Your first experiments may be frustrating. Some will fail. You will also discover tasks where your existing way of working is still better.
But do not confuse being a beginner with being incapable.
The professionals who are best prepared for what comes next may not be the ones who master every new AI tool. They may simply be the ones willing to stay in the learning curve long enough to discover what is worth learning.
And if fear, uncertainty or resistance is making it difficult to determine how AI fits into your career, that may be worth examining more closely. Career coaching can help you identify the beliefs and behaviors that may be keeping you stuck, clarify what you can control and build the adaptability to navigate what comes next.
Learn more about Career Coaching