Dylan Dupuis

Marie Johansen

@greenfrog499172 · Jan 12, 2026 4:34 am

Am I the only one who finds Microsoft Copilot painfully behind?

I really wanted to like it. It’s built into Windows, it’s free, and Microsoft is throwing everything at AI. But after giving Copilot a solid try for the last few months, I’ve come to a frustrating conclusion: it feels like it’s a good 12 months behind the curve compared to models like ChatGPT, Claude, Perplexity and Gemini.
My main gripes:
* **The “Helpfulness” Filter is Aggressive to a Fault:** I ask for a slightly creative or edgy rewrite of an email, and it falls over itself with “I can’t assist with that.” I’m not asking for anything crazy! Other models understand nuance and intent way better.
* **Output is Just… Weaker:** The responses often feel generic, shorter, and lack the depth or insightful “spark” I get elsewhere. It’s like talking to a very cautious, middle-management AI.
* **Context Gets Lost:** I’ll have a back-and-forth and it seems to forget the core of what we’re discussing way faster than its competitors. The conversation threading feels brittle.
* **Integration is Its Only Win:** Sure, pulling data from my PC or summarizing a PDF in Edge is neat, but if the core brain isn’t as capable, the fancy integrations feel like a faster horse and carriage when eve