I’ve finally decided to pull the chord and give this whole AI stuff a real college try, as they say.
After some consideration, I used the highly scientific Eeny-Meeny-Miny-Moe method and landed on Gemini.
So how’s it going?
Images
Kids love it. The colouring pages are fun for them. Even the neighbours kids are coming over asking.
Though I find myself really needing to ask “can you try it again, but this time….” comments.
I’ve also been struggling a bit on dimensions. While the images look nice, I usually want them to be specific in shape, and it’s been a struggle to say the least.
I really wonder the long books length prompts that are required for some of the more advanced uses of these tools.
Coding Agents
When I first started trying to code with Gemini, it was a small pain because I essentially was given step by step prompts on what to do and samples. Meh…
Cursor was fun… but felt like it added a bit of bloat – lot’s of code that maybe didn’t need to be there.
Then game Antigravity
It cleaned up after itself and I though… well that’s interesting.
The way it does it’s briefs and summaries I find very intuitive to read through through the steps and thoughts the agent explored.
I gotta say, it makes some of the lighter work easier when you can give it the simpler tasks and it just does it. Give a quick read to make sure it went through the right mental steps as checks and balances.
I haven’t fully flexed it’s muscles, but I think I just might and see what comes of it.
Notebook LM
I made a separate post in this… the podcast feature blew my mind a little.
I don’t have any other idea’s that need that deep of a data set, but if I did – that’s a cool tool.
Gems
Gemini has Gems, there are like little helper agents. I tried it for a few days for a bit for things like productivity and ADHD exploration.
Maybe it’s what I tried to have them help me with? However my overall take is that it would be so easy to have it hold my hand and tell me what to do next. But, that’s kinda one of those things that has my dooms day yellow flag going off a tad.
Overall Thoughts
I gotta say, this stuff is fun, and enticing. It’s so easy to see the time savers. I think I’ll keep playing for a bit more, however I’ll probably push it a little further out to an arms distance…
Just one full week in, and I can quickly see and even feel a little bit of the ethical pitfalls. Those psychological cracks it the mind that it can easily slip in through.
If not careful. If not resilient.
Being told that every question I have is a great question; that I’m always making great progress; that I am awesome. Doing it over and over, feels off. Like surrounding yourself with digital “yes” folk.
Giving up my own agency and asking it what I think I should do next. Maybe use its take on a subject that perhaps I should have my own take on. Perhaps the right course in life would be to make a left turn, zig when I should zag. If I’m not careful enough I might give it enough of my agency it will tell me to keep driving straight, and I’ll do it, even though into the proverbial lake straight ahead.
Changing how I communicate. How I ask questions or better yet, how I “prompt”. The novels I would write for it, to do the things I want it to. And it will get close enough. Though would I even give the same detailed prompts for the real people around me. I shouldn’t need to. They are humans and should just get it. The AI is a computer so it needs me to be more considerate than to those around me. It needs me to be more detailed because of it’s mind.
While this might feel like a bleak outlook, I like to sit in the grey. I can still see the good. I thing the ethical conundrums of our technology have up till now felt pretty small in comparison. This one I can quickly see and even feel the weight.
Though I’m just spit ballin’ – the good old classic “If I had more time I would have written a shorter [post]”
Oh who was it who said that?
“Hey Google…”
