Hello,
I need to talk about what just happened. Most of the coverage is about politics, ethics, and corporate drama. Nobody is talking about what this actually means for people like us, professionals trying to use AI to work smarter.
Here is the timeline.
February 9th. OpenAI started showing ads inside ChatGPT. Not on the side of the page. Inside your conversation. On your very first message. Expedia, Target, Ford, right there in the chat. If you are on the free version or the 8 pounds/month Go plan, you are seeing ads. On Reddit, 68% of comments were negative. On X, posts about it got over 10 million views, mostly critical.
February 28th. OpenAI signed a deal with the US Department of Defence to deploy AI on classified military networks. The same week, Anthropic (the company behind Claude) publicly refused. Their CEO wrote that he could not in good conscience agree to the Pentagon's request. Anthropic specifically refused to allow its AI for mass surveillance or fully autonomous weapons. OpenAI took the deal.
March 1st. #QuitGPT went viral. 1.5 million people reportedly took action. 700,000 users actively left.
March 2nd. Claude hit #1 on the Apple App Store. Signups up 60% since January. Paid subscriptions more than doubled. Then their servers crashed. They could not handle the flood.
13 days. That is how fast things moved.
The Part Nobody Is Talking About
I watched professionals in my network, people who have been using ChatGPT for months, post things like:
"So if I cancel ChatGPT... what do I use instead?"
"Does Gemini do the same things? Do my prompts work there?"
And the honest answer for most of them is: they do not know.
Because they did not learn AI. They learned ChatGPT.
They know the layout. They know where to click. They might have custom GPTs set up. Maybe they use the memory feature.
But none of that transfers. Custom GPTs do not move to Claude. Your ChatGPT memory does not carry over. The interface quirks you got comfortable with look completely different on another platform.
Most professionals did not build an AI skill. They built a ChatGPT habit.
A habit means your whole AI experience lives inside one app. You open it because that is what you always do. You type something, hope it works, and rephrase if it does not.
A skill means you know how to structure a prompt that works on any platform. You have a prompt library saved somewhere you control, a Google Doc, a Notion page, and a folder on your desktop. You have identified the specific tasks in your week where AI saves you time. And your system is not tied to any single tool.
The people with a skill are not panicking this week. They are copying their prompts into Claude, making small adjustments, and carrying on. Their system belongs to them, not to OpenAI.
What You Should Actually Do This Week
1. Export your best work before you cancel anything. If you have good prompts, templates, or strategies refined inside ChatGPT, save them in a document that lives outside the platform. People are cancelling in frustration and losing everything they have built. Capture it first.
2. Know the differences. Claude tends to be stronger in writing, nuanced analysis, and following complex instructions. Gemini is tightly connected to the Google ecosystem and has a massive context window for large documents. ChatGPT is still strong for structured reasoning and has the largest plugin ecosystem. Your approach should work across all of them.
3. Build your prompt library outside any AI tool. This is the most important thing you can do this week. Pull your best prompts out. Organise them by task. Make them yours. Your prompt library should live in a place you control.
4. Learn the skill, not the app. The tool landscape will keep changing. If your AI knowledge is "I know how to use ChatGPT," you will start over every time something shifts. If your AI knowledge is "I know how to structure prompts, give context, and build a repeatable workflow for my role", you will never start over again.
AI, Done Properly, Starts March 16th
I built this course for exactly this moment.
Not a ChatGPT course. A course that teaches you to build a system that works anywhere and belongs to you.
Weeks 1-2: We audit your real working week and identify the 5 tasks where AI saves you the most time.
Weeks 3-4: We build your personal prompt library. Tested, refined prompts that work on ChatGPT, Claude, Gemini, wherever you choose. Saved somewhere that belongs to you.
Weeks 5-6: We turn it into a weekly workflow. Repeatable, Monday-to-Friday, built around your schedule.
By the end, you do not have a ChatGPT habit. You have an AI skill that goes wherever you go.
10-15 people. Small on purpose, I work with everyone individually.
£297 for 6 weeks.
Reply to this email if you are interested. No sales call. Just a conversation. You can also find more details here: https://configurai.com/
And if you are not ready, this newsletter lands every Wednesday either way. Free. No strings.
If someone in your life keeps saying, "I really should learn how to use AI properly", forward them this. Sometimes the right nudge is all it takes. https://configurai.com/
Thanks for reading,
See you next Wednesday with more ways to use AI without losing your mind (or your credibility).
Orgesa Meli
P.S. If this saved you from a future hallucination disaster, forward it to someone who's using ChatGPT for proposals, reports, or client work. They'll thank you later. Subscribe to my community here.
