Artificial Intelligence

Artificial Intelligence is a misnomer.  AI is a poorly used phrase to explain machines that learn by processing external inputs and algorithms recursively.

There isn't anything really Artificially Intelligent about it, computers are just machines. Clever software written for these machines can make modifications to a machines processing cycle and algorithms over time. Dependent upon computed results or additional information provided by outside sources, this software can make a machine appear to be intelligent.

Using the misnomer definition of AI, plenty of things about KYNGIN could be considered artificially intelligent.  KYNGIN adjusts process priority values based on demand and usage, it throttles content delivery when environments are being attacked or over subscription events take place, it stops spam from hitting servers when it's been determined the source of the spam has recently started sending UBE, etc. (BTW, we don't call KYNGIN AI, because it's now officially a marketing term that fuels FOMO and serves minimal purpose aside from manipulating the ignorant. Most companies using this term are predatory.)

Fear not, as none of this "AI" is intelligent in any conscious sense.  It's only a series of algorithms used by the machine to learn, and then the machines use algorithms to react to the variables that have been provided to them by outside sources.  It's not intelligent, it's just a machine, and we won't use the term AI here to describe KYNGIN because by doing so would be a pretty clear red-flag indicator to technology world that we don't really understand the field all that well. AI is yet another buzz word, largely capitalized on by companies without their own innovation, or intentionally due to bad behavior.

Addendum 2026

Years have passed since the previous writeup. AI still leads the way as one of the most over valued technology processes in our history. Real world use cases seem to be friendly (read "sycophantic") chat bots, voice to text transcription services, textual summary generators, data sorting and organization tools, and a whole slew of content generation (copyright theft?) tools in the form of computer code, audio, video, and images.

The real world productive use cases seem to be minimal, yet the hype for this has fueled economic instability for the last couple years with many people at this stage expecting the bubble to burst soon enough. Countless stories of business leaders telling their employees to adopt AI (likely because chatbots work well for the narcissists that like to promote it, but also because many business leaders are the first to buy into any sales gimmick that can give them the competitive edge) without seeing the security and privacy consequences hidden in plain sight. While the naive technology workforces not knowing any better adopt where told, the skilled workforces stay resistant, seemingly not seeing the value.

We don't use any forms of AI here at Mindpack Studios, and still wont use them as a sales gimmick for KYNGIN or any other product we build. We've spent some time testing use cases, largely using local LLM's (trusting any form of public cloud service is risky, most of these AI companies will eventually resort to mining all your personal data to sell you things you don't need years later, or helping you choose your elected officials that are in the best interest of the cloud company itself - which all measures out to be a personal security reduction, so we stay away).

We've located some viable use cases for AI, possibly not the generative parts which are largely fraudulent, but the problem solving, code inspection, data summary portions of AI (a kinda hyper grep/awk/encyclopedia). It's probably not worth the cost of burning the world with how much energy these tools seem to take, but humans have a hard time seeing the costs when the tailpipe lacks visible emissions (i.e. EVs still burn lots of coal). And possibly some day this will resolve to smaller more efficient compute systems that deliver better energy consumption numbers.

But overall, in our testing with a perspective amassed from a few decades of technology experience, it simply all just gets in the way of our workflow. Simply, we deliver quality services without help, and we take too much pride in our development. AI code and documentation generation does work, it's just not the greatest quality, and brings with it a bit of icky-ness as you feel like the products you've built aren't really yours. A bit like winning a board game by cheating, fun the first time, for about 11 seconds. The cost in pride loss by basically cheating doesn't work out well for us or our clients, and I believe this costs far more than any front-end time savings (which you'll likely lose with age) we'd make by using AI code or documentation generation.

Instead of AI adoption, we'll keep generating code and participating in technology the old fashioned way: Spending every day improving our quality, learning from our mistakes, reading lots of documentation, and having fun. Oh, and waiting for this bubble to burst (and get out of the way), so we can buy technology components again at reasonable prices. :|