Why I Stopped Using ChatGPT (And How You Can Too)

Recently, I extracted my data from ChatGPT and deleted my account to search for other AI providers. You may be wondering why I would do this, ChatGPT was one of the first popular AI LLMs (Large Language Model), and I was one of their first 3% of customers. ChatGPT was always known for pushing the envelope of AI technologies, but (at least in my opinion) not anymore.

ChatGPT was known for being the first widely-used LLM, reaching 1 million users within 5 days after its release. At the time, there really was no standard for a good LLM, so ChatGPT set the bar, but that bar, compared to other LLM’s today, was very low.

I and many others have had many issues with ChatGPT. ChatGPT never tells you that you are wrong, or that something you want to do is a bad idea. ChatGPT never tells you that it doesn’t know something, and is willing to lie about information to satisfy you. ChatGPT seems to have somewhat of an attitude. ChatGPT always gives unnecessarily lengthy responses. And ChatGPT seems very corporate and cold.

The final nail in the coffin for me was when OpenAI (the company behind ChatGPT) made a deal with the Department of War “to accelerate operational tempo, enhance decision-making superiority, and address critical national security challenges.” This made me want to switch to another AI even more! I don’t want to use an AI that works for the government! And even worse, I don’t want to use an AI that has the power to bomb a country!

After hearing about this deal, I decided to find a new AI to migrate to, I thought, “Why don’t I switch to Claude, the AI that the department of war originally wanted to use. Claude said no due to privacy reasons.” And turns out, two of my favorite youtubers: CarterPCs and Nathan Espinoza, switched to Claude by Anthropic, so I tried it.

Claude is great, I would say it’s definitely better than ChatGPT! The UI is much more warm and cozy. Claude is willing to tell you when you’re wrong, when it doesn’t know something, or when something you want to do is a bad idea. Claude gives shorter and more simple responses (unless you specifically ask for a longer response). It’s easier to tell Claude your preferences about its responses. And also, Claude has a really great artifact feature, which are extremely customizable mini apps such as “Anthropic Office Simulator,” “Molecule Studio,” “QR Code Generator,” “Your Life In Weeks,” “Forest Explorer,” and “Mood Palette.” 

So, you may be asking “How can I switch to Claude?” A good (and official) instruction guide on how to switch to Claude is located at this link: 

https://claude.com/import-memory 

But here are some basic instructions. First, you paste the following prompt into whatever AI LLM you use: I'm moving to another service and need to export my data. List every memory you have stored about me, as well as any context you've learned about me from past conversations. Output everything in a single code block so I can easily copy it. Format each entry as: [date saved, if available] - memory content. Make sure to cover all of the following — preserve my words verbatim where possible: Instructions I've given you about how to respond (tone, format, style, 'always do X', 'never do Y'). Personal details: name, location, job, family, interests. Projects, goals, and recurring topics. Tools, languages, and frameworks I use. Preferences and corrections I've made to your behavior. Any other stored context not covered above. Do not summarize, group, or omit any entries. After the code block, confirm whether that is the complete set or if any remain.

Then you copy and paste the results into Claude’s memory settings. That’s it! Claude will update its memory automatically. So I hope this article was helpful and informative about Claude. Thanks for reading!