The conversation around artificial intelligence has moved quickly. What recently felt experimental is now embedded in everyday workflows, from writing and research to analysis and decision support. A number of well-known companies have emerged as clear leaders and, naturally, discussion has followed around which systems perform best for different tasks.
As adoption has grown, so too has the tendency to compare tools. People share preferences, highlight strengths, and discuss which system produces better writing, clearer thinking, or more reliable outputs. These discussions have some validity, because humans interpret outputs subjectively, and patterns of preference tend to form over time.
General consensus, however, does not necessarily tell the whole story.
The overlooked variable: user maturity
What is often missing from these comparisons is the role of the user. As individuals spend more time working with AI, they develop a better sense of how to guide it, structure requests, and refine outputs. We don’t always notice it, but it has a direct effect on the results.
When someone first uses AI, they typically engage at only a basic level. Prompts are simple, expectations are unclear, and the interaction is largely reactive. Over time, this changes. Prompts become more structured, intent becomes sharper, and the user begins to shape the response rather than simply receive it.
Why second impressions feel stronger
This creates an interesting dynamic. A user might begin with one system, using it in a relatively unrefined way. Later, they try a different system, but now with more experience, more curiosity, and a greater willingness to explore advanced features. The second system appears more capable, but the improvement may be coming from the user, not the tool.
In many cases, the first system had similar capabilities all along…but they were not being fully exploited.
The compounding effect of experience
As people move between systems, their level of sophistication tends to increase. They test more nuanced instructions, experiment with tone and structure, and push the boundaries of what the tool can do. Each new system benefits from the experience gained with the previous one, creating a sense of continuous improvement that may not be entirely attributable to the technology itself.
This leads to a skewed comparison. What feels like a leap in capability may actually be — at least in part — a reflection of accumulated user skill.
Improved versions of the same system
There is another layer to consider. These systems are evolving rapidly, with frequent updates and improvements. Comparing one system at an earlier stage to another at a more advanced stage can create misleading conclusions. A tool used months ago may now perform very differently, yet the original impression often lingers.
Convergence and adaptability
It is also worth noting that many modern AI systems share underlying principles and capabilities. With well-structured prompting, it is often possible to guide different systems toward very similar outputs. While they may vary in style or default behaviour, the gap can narrow significantly when the user provides clear direction.
With expert prompting, differences between systems are sometimes less fixed than they appear.
A more grounded perspective
Most people have been on some kind of journey with AI, whether they realise it or not. Some have leaned into it quickly, others have held back, and many are still figuring out where they stand. What tends to receive less attention is what happens after that point, as people become more capable and more deliberate in how they use these tools.
For those exploring types of artificial intelligence, it is useful to look beyond surface-level comparisons and consider the role that personal development plays in using it. The effectiveness of any system is closely tied to how it is used (how a car performs, for example). And that relationship becomes more pronounced over time. As users become more capable, the tools appear to improve alongside them.
It may be worth revisiting earlier conclusions. The question is not simply which system is better, but how your own approach has evolved alongside them.