This told essay is based on a conversation with Sidhant Bendre, 26, co-founder of Oleve, an AI-driven consumer software portfolio company based in New York. His words have been edited for length and clarity.
I used ChatGPT for about two years, then moved to Claude as my preferred model and ended my company’s ChatGPT subscription.
When we started our company, Oleve, we were already using AI, so we’ve always relied heavily on it in our workflows, from coding to marketing to the hiring process. ChatGPT has been our go-to model. Things changed after Anthropic released a 4.5 model suite for Claude last fall.
We’re not being turned away from ChatGPT. We were drawn in by Claude.
From the beginning, Claude generated code with fewer bugs than ChatGPT.
I didn’t have any major complaints with ChatGPT that forced me to switch, but I think one of the promises of AI is to speed up your work. Thanks to Claude, I can now work faster because I spend less time fixing things.
When I was primarily using ChatGPT, I noticed that the content it generated was not very human-like. I ended up using too many emojis in places I wouldn’t, and it felt really forced. I haven’t strictly used ChatGPT for a while, but when I go back to testing prompts using the model, I still get redundant responses.
Claude is good at imitating human sentences. Even before Claude’s 4.5 suite came out, I was listening to students discuss how this model was very good at imitating their writing style after giving them a sample.
Claude’s coding allowed us to automate much of our development time using the blueprints we already had. Now you can focus on the product itself instead of spending a lot of energy on build time.
The biggest difference is how the model understands nuances
The biggest difference between the models seems to be parsing context and considering nuances.
If I give Claude a large research document and I’m looking for something small, Claude almost always knows when to be concise, provides the right context, and highlights what I want more detail on.
At times, it felt like ChatGPT was overcompensating for bad decisions about what to focus on by spewing out more content. If you’re currently using it, you usually have to ask it multiple times to summarize it or redirect it to be more concise.
Since switching to Claude, we’ve integrated AI into more of our workflows. If you know it’s worth it, it makes more sense to invest the necessary time and energy into it.
Claude gave us our own problems
Lately, Claude has been giving us his own problems with bugs. Some of my messages and chats are disappearing and it’s frustrating.
It’s not perfect. My trust in Claude is a bit broken, as I feel like the model’s output still has some hallucinations at times. Even if you use web search with the new Opus 4.6 model, you will need to push back to self-correct these scenarios.
I’m still a big fan of Claude’s suite, but I think I’ll switch if something better comes along
I wasn’t really paying attention to the larger OpenAI vs. humans debate. If ChatGPT launches a suite of products that provides value that is important to me, I will give it a try. I’m not so attached to Claude that I don’t want to try anything else.
I think the biggest thing for me is that Claude feels more like what you’d expect from an AI. Each model has brought great value, but I feel like it has freed me from the daily grind and given me back time to think about the big picture.
Do you have a story to share about the use of AI? Contact this reporter, Agnes Applegate: aapplegate@businessinsider.com.
