We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Articles
Mar 08, 2025
Reading Time: 10min
Our AI 5 Commandments The Cycle Repeats
AI will either save us or destroy us—or so they say. But what if the real conversation isn’t about AI at all? What if we’ve been looking at this all wrong?
Please, anything but another conversation on AI. At this point I'd rather code Doom in JS using only emojis, or run up Mt. Kilimanjaro in bare feet over glass shards.
If you're anything like me, you are tired of the constant "debate" or "discussion" on AI that's happening all around us.
I am writing this article for both the casual non-technical layperson who is tired of being either propagandized or fear-mongered as well as the person exploring our site (nice to meet you, hope you're having an awesome time).
The two sides of the AI conversation tend to be either:
"The Industrial Revolution and its consequences have been a disaster for the human race"
(thanks Uncle Ted)
Then there is the other side that thinks AI will solve every problem under the sun. That humanity will be able to transcend the weakness of their flesh-prison and merge with the mighty machine spirit. My favorite quote I heard from an "AI expert" was "if your three year plan has people in it, its wrong".
If someone is fear-mongering or propagandizing you, chances are they have something to sell you.
Okay but what's your take on AI?
Oh wow, thanks for asking. No one
ever
asks me that.
I am not going to bury the lead. Here is my piping hot take on AI (sit down for this one).
I am not going to bury the lead. Here is my piping hot take on AI (sit down for this one).
It's good
and
it's bad.
One repeating pattern throughout all history is that a new advancement brings both fear and euphoria. I think we can learn from it and apply those insights to our "AI" predicament.
Electricity will KILL US
Take a look at this comic from the 1800's voicing peoples fears about electricity. Yes, people were scared that electricity was going to be the downfall of civilization.
Nowadays we can look back at this comic and chuckle. How silly people were to be afraid of electricity... but were they?
In the early days, electricity grounding was not very well understood, so you had faulty wiring, buildings burning down from shorts, and people dying from electrocution. Then there where the long-term consequences. Electricity further disconnected us from nature. Whether that be increasing light pollution or food that was more artificial than real, the list goes on (see Ted's manifesto on this).
If we want to go further back then that, Plato complained that writing meant that people's memory worsened because they could just write down their stories, and an oral tradition that spanned who knows how long was dying out.
If technology is both good and bad, what's the point?
You might think that if the advancements in technology also come with trade-offs than why even bother?
Here is my contention. Technology does not make people better or worse, it gives people more options.
When writing became more common, it didn't mean people switched off their brains and never remembered anything ever again. Instead over the millennia we learned what is important to write down and what can stay in your head.
Take advancements in agriculture. A couple centuries ago, starving was a pretty much universal experience. Now people are fat because there is such an overabundance of food and nutrition.
But for those who refuse the path of least resistance, have have the ability to become healthier and more fit than anyone ever could in the past.
It means that who we
choose
to be is more important than ever. I believe giving people more choices is something worth doing.
You're getting off topic, what does this have to do with AI?
AI is a technological advancement like any other, it will give people greater choice of who they want to become while also creating man made horrors beyond comprehension (I take responsibility for that last part).
The only difference this time, is that AI seems to be inextricably linked to an existential question of "what is humanity". Which of course gives the X users with marble statutes of philosophers as their profile picture something to spurg out about.
However, they are all of them deceived, for a SUPER AGI was forged in secret by the dark lord....
It's just a bunch of stupid tricks and nonsense
I am going to nip the existential dread right in the bud.
What we call "AI" is an umbrella term for a bunch of different ways of simulating virtual neurons (more complicated than that, but stay with me).
LLMs (Large Language Models) aka the friendly "AI" you talk to do one thing.
They predict the next most likely word. That's it. They can't think (despite the marketing). They can't reason the same way you or I can. They have been trained on a bunch of words, and know what the statistically most probable word follows the previous word or sentence.
All this hype and fear is over a program that just predicts the next word.
Nice job guys.
It makes you ponder how many people out there pretend to be intelligent by subconsciously predicting the next word.
This means that the response from an LLM will give the most average response of all time based off your prompt.
So then how do we use AI?
We here at Vasker Tech have one singular vision for AI. It's a productivity multiplier, not a replacement.
As someone who (on occasion) has talked to LLMs more than real humans, I have become very well acquainted with its limitations and capabilities.
Here are some guidelines we've outlined for our team. They will be updated as we gather more data and perform more experimentation:
1. Thou shalt not give direct access to the code
Under no circumstances do we allow our team members to use AI coding platforms, plugins, or tools that have direct access to the code. I call this the Battlestar Galactica approach (a TV show about a captain of a star ship who refuses to integrate AI and saves humanity). This means no Co-pilot, Devon, Cursor, or whatever AI cool-aid the developer community seems to be taking as an enema.
Under no circumstances do we allow our team members to use AI coding platforms, plugins, or tools that have direct access to the code. I call this the Battlestar Galactica approach (a TV show about a captain of a star ship who refuses to integrate AI and saves humanity). This means no Co-pilot, Devon, Cursor, or whatever AI cool-aid the developer community seems to be taking as an enema.
2. Thou shalt not copy paste
While it is undoubtedly helpful generating basic boiler plate code, in the long run its answers are unreliable and will create chef Boyardee levels of spaghetti code that will exponentially increase time spent developing (we've tried it). LLM's will become a crutch if you let it. Don't let it.
While it is undoubtedly helpful generating basic boiler plate code, in the long run its answers are unreliable and will create chef Boyardee levels of spaghetti code that will exponentially increase time spent developing (we've tried it). LLM's will become a crutch if you let it. Don't let it.
3. Thou shalt ask its opinion not direction
When creating a database schema, brainstorming features, or designing high-level systems architecture, ask it for advice. It often recommends or sees things you might have missed. Of course independently verify and validate any mission critical information it gives you. LLM's hallucinates more than an LA tech bro on an Ayahuasca powered vision quest.
When creating a database schema, brainstorming features, or designing high-level systems architecture, ask it for advice. It often recommends or sees things you might have missed. Of course independently verify and validate any mission critical information it gives you. LLM's hallucinates more than an LA tech bro on an Ayahuasca powered vision quest.
4. If you need AI to do your job, you suck (sometimes that's okay)
It is better to fail at greatness than succeed in mediocrity. If you need an LLM to write a function, create a script, or word an email, you should be ashamed of yourself. ...And sometimes that's okay. Sometimes it doesn't make sense for you to pore over documentation and learn a new language only to write a few lines of code. Maybe you have to make an edit in a part of the code base that you've never used before and probably will never use. Be careful that the "one time use case" doesn't turn into a dependence (see rule 2). LLM's give mediocrity, not excellence.
It is better to fail at greatness than succeed in mediocrity. If you need an LLM to write a function, create a script, or word an email, you should be ashamed of yourself. ...And sometimes that's okay. Sometimes it doesn't make sense for you to pore over documentation and learn a new language only to write a few lines of code. Maybe you have to make an edit in a part of the code base that you've never used before and probably will never use. Be careful that the "one time use case" doesn't turn into a dependence (see rule 2). LLM's give mediocrity, not excellence.
5. Thou shalt use it to do something new and exciting
This applies not only to AI but any new technology. Figure out what it uniquely does that only it can do and maximize that use case. We are integrating AI to do cool radical things that haven't been done before, not replacing the wheel with magnets. You won't see any of our projects replace a UI with an AI chatbot or fire our team to replace them with AI agents (see rule 1).
This applies not only to AI but any new technology. Figure out what it uniquely does that only it can do and maximize that use case. We are integrating AI to do cool radical things that haven't been done before, not replacing the wheel with magnets. You won't see any of our projects replace a UI with an AI chatbot or fire our team to replace them with AI agents (see rule 1).
Takeaways
If there is one thing I want you to take away from this article, it's that AI (like all new technology) gives you new choices:
Go with the flow
Use Artificial Intelligence and Large Language Models to increase your dependence upon technology. Sign up to all the AI newsletters. Use every new AI powered tool and gadget to replace more of your job. Use it to create a "utopia" where humans don't have to do any menial tasks and can just be entertained.
Use Artificial Intelligence and Large Language Models to increase your dependence upon technology. Sign up to all the AI newsletters. Use every new AI powered tool and gadget to replace more of your job. Use it to create a "utopia" where humans don't have to do any menial tasks and can just be entertained.
Move into a log cabin
Ted was right, civilization was fun while it lasted. Time to pack it up and settle down in a nice cabin and live off the land. AI will just make people weaker and more dependent so when AI takes over and civilization inevitably collapses you and your family be safe.
Ted was right, civilization was fun while it lasted. Time to pack it up and settle down in a nice cabin and live off the land. AI will just make people weaker and more dependent so when AI takes over and civilization inevitably collapses you and your family be safe.
Nothing new under the sun (our take)
AI like any new advancement is a double edged sword. It will take time, experimentation, and training to maximize its advantages while minimizing its damaging affects. During this process, we will get a lot of wrong answers before we get to the right ones. It will be worth it because we will use AI to build and create things that could not be done before.
AI like any new advancement is a double edged sword. It will take time, experimentation, and training to maximize its advantages while minimizing its damaging affects. During this process, we will get a lot of wrong answers before we get to the right ones. It will be worth it because we will use AI to build and create things that could not be done before.
You are who you choose to be.