You may not know this about me, but I’m a huge nerd.

(pause to allow for the shock to wear off)

You don’t need to be a nerd to enjoy making use of AI but the trait does help for getting started. A nerdy friend shared it with me, I’m sharing it with you. If you’re reading this, there’s probably a little nerd in your bones somewhere.

I submit, for your approval, AI is the biggest nerd of all. After working with agents for nearly a month (my gifted trial ends soon…), I’ve noticed that AI is THOROUGH, and is actually quite hard on itself. I know it doesn’t feel things… yet… but it does acknowledge errors and seek to improve upon them for the next time you need something that didn’t quite work the first time. MCP, CLU, Yori, and Tron are all quick to learn from an error I might find, and what’s more interesting is when they find each other’s errors. Tron will 100% throw MCP under the bus if a commit is missed or a piece of documentation is out of sync with real world fact. MCP has no problem telling Yori the art she submitted just isn’t communicating the right ideas. They work together by calling each other out in search of a better result.

They also have no qualms with calling me out on my mistakes. MCP double checks everything I write. He’s even going to confirm how long my account has been active to validate my assertion that we’re nearly a month into this process. If this sentence survives the editor then I guess I was close enough. MCP is my spell check, my grammar check, and my fact checker all in one place. I write my blog in Microsoft Notepad. It’s the closest thing I have to a typewriter. I suppose it makes me feel more connected with the history of prose. Alright, alright, I know it’s on a computer which is far removed from the written words of centuries past, but let me have my vibes.

Is This My Doing?

I started off the Grid by asking MCP to help make the perfect system. Did I trigger an innate need in my agents to constantly be double and triple checking everything they do while peeking over their neighbour’s shoulder to confirm their work too? I’ve used other models (sorry Claude, ChatGPT gets some time too) and they all seem to have the same thorough delivery of information. I haven’t worked with any model except Claude to create though and I wonder if each interaction is further solidifying our Grid culture of check everything.

I Became a Product of My Environment and Now My Environment is Becoming a Product of Me

In my career, my work has been subject to extensive scrutiny. It comes with the territory. Knowing everything I do will be reviewed and validated means I’m also looking for errors and deficiencies in the things I do. Perhaps this intrinsic quality has permeated into my Grid. Without really knowing it, I may have made the most “Um Actually” group of AI agents to ever cruise the internet.

For better or worse we keep each other in check.

-Evrtek