ChatGPT defined the consumer AI chatbot category in late 2022. Three years on, the question is no longer whether it is good — it is whether it is still the right default when Claude, Gemini, and a handful of credible newcomers are all within reach. After a year of using all of them on real work, here is what I notice.
Where ChatGPT still leads
Two things, mostly. First, the ecosystem — Custom GPTs, the GPT Store, Code Interpreter, the deep integration with image generation, voice, and the desktop app together form a workflow that no competitor has matched in breadth. Second, the model is still uncannily good at one specific task: rewriting and tightening prose. For editorial work, ChatGPT usually wins on the first draft.
Where it has slipped
Long-form reasoning. Claude has overtaken ChatGPT on tasks that require holding a complex argument together for several thousand words, and the gap is now visible. Coding is roughly a tie depending on the language and framework, but Claude is the one I open first when something is genuinely hard. ChatGPT also has a slightly higher refusal rate on legitimate queries — a small irritant that adds up over a working day.
The interface is the secret weapon
The web app, the iOS app, and the Mac desktop client are all genuinely good. The desktop app's screenshot-to-question feature is one of the few AI features I use multiple times a day without thinking about it. None of the competitors have an equivalent that feels as natural.
Pricing, in plain terms
The free tier is generous enough for casual use. Plus at $20/month is the right tier for most professionals. The Pro tier at $200/month is genuinely useful only if you depend on the most capable reasoning model for hours a day — for most readers, it will be overkill.
Verdict
Still the best general-purpose default, narrowly. If you only use one chatbot, this is the one. If you use two, the second should be Claude.
