Throughout my career, I’ve seen many things change. I’ve gone from Borland to Visual Studio, from vi to Sublime Text, from Sublime to VS Code… And believe me, each change was a deliberate decision that cost me weeks of adaptation. But what’s happening now with AI tools is something completely different.
I’ve found myself using Copilot in the morning, trying Cursor in the afternoon, and checking out Claude Code before going to bed. And I’m not alone. Developers have gone from being faithful as dogs to our tools to being… well, promiscuous.
The End of Two Unbreakable Dogmas
Recently I read a RedMonk article that made me reflect on this. Historically, there were two things we took for granted in the development world:
First: Developers were loyal to their tools. Once you mastered Emacs, why change? Once you got used to your favorite IDE’s keyboard shortcuts, why learn others? It was the ugly duckling syndrome in its maximum expression.
Second: Developer tools were free. Period. We paid for everything except our code editor, our compiler, our debugger. It was unthinkable for a company to ask for budget for something you could download for free.
Then came 2021 and GitHub Copilot changed the rules of the game.
When Microsoft Broke the Mold
Copilot was revolutionary not just because of the AI, but because it proved that developers WERE willing to pay for tools. In two years it reached $100 million in recurring revenue. Can you imagine? An industry that had resisted for decades paying for development tools suddenly opened their wallets.
But what was most surprising wasn’t that. What was surprising was that Cursor came along and in 12 months achieved what Copilot did in two years. And many of its users were… former Copilot users.
The Cambrian Explosion of Tools
Right now we’re experiencing a Cambrian explosion of AI development tools. Just to give you an idea, in the last year alone: Cursor, Cline, Windsurf, Bolt, Replit, v0, Lovable, Same.dev, vibes.diy… And each has its particular approach.
Some are more conversational, others more specialized. Some give you total control, others do everything for you. Some maintain the traditional IDE interface, others are little more than a text field.
And here’s the curious thing: developers are trying them all.
Why Are We So Unfaithful Now?
After much thought, I think there are several reasons:
AI changed the rules of the game. When the main input is natural language instead of code, keyboard shortcuts and muscle memory matter much less. Learning a new tool no longer means relearning 200 key combinations.
Each tool is genuinely different. It’s not like before, where switching from Eclipse to NetBeans was basically the same with buttons in different places. Cursor works one way, Cline in a completely different way.
Token limits force us to be creative. When you run out of credits in one tool, you simply jump to another. It’s like having several credit cards for when one hits the limit.
Experimentation is part of the process. We’re in a phase where no one knows which approach will win. It’s like the early days of the web, when no one knew if HTML, Flash, or something completely different would triumph.
The Personal Experience
In my case, I’ve tried at least 8 different tools in the last 6 months. Each has its place:
- Copilot for day-to-day in VS Code
- Cursor when I need longer conversations about code
- Claude for architecture and planning
- v0 for quick interface prototyping
And yes, I switch between them depending on the project, the time of day, or even my mood. It’s something I would have considered crazy five years ago.
The Economic Side
Here’s what worries me: the cost. I know developers who are spending hundreds of euros a month on AI tools. This is creating a division that didn’t exist before.
Before, a junior developer had access to exactly the same tools as a senior. Now, whoever can afford better AI tools has a real advantage. It’s as if suddenly purchasing power determined your productivity as a developer.
Reflections for the Future
As someone who has seen many technology cycles, I can tell you this is temporary. The industry will consolidate, some tools will disappear, others will become dominant.
But meanwhile, my advice is:
Don’t marry a tool. For the first time in decades, being promiscuous is an advantage.
Learn the concepts, not the tools. Understand what each type of tool does, not how to use every button.
Maintain a budget. It’s easy to accumulate 10 subscriptions of 20€/month without realizing.
Document your process. When you change tools (not if, when), having documented your workflow will save you time.
Conclusion
We’re living a unique moment in the history of software development. For the first time in decades, loyalty to tools is a disadvantage. Promiscuity has become survival.
As a developer with experience in multiple technology cycles, I can tell you it’s exciting and exhausting at the same time. But it’s also a unique opportunity to redefine how we work.
The next time you find yourself testing the 15th AI tool of the week, don’t feel guilty. It’s part of the process. Just make sure all that experimentation is making you a better developer, not just busier.
What do you think? Are you one of those who stays faithful to one tool or have you also succumbed to technological promiscuity? I’ll read you in the comments.