Over the last few years, the world of AI has boomed into a global phenomenon. We cannot pick up a book or open an article on the internet without being reminded of how AI has positively impacted our lives. Personally, I have followed each new press release with excitement and poured through hundreds of hours of video footage showing the amazing results that can be obtained from the various AI models available for use.
And I have bought into the hype that we are all benefitting from this surge in artificial intelligence. I have to wonder though, whether it has been for the better.
Let me explain why.
I recently came to the following realisation when I was performing a menial task at home, putting away my clean laundry. As you can imagine this is not something that I enjoy doing. In fact, I would rather do anything else. And there are many tasks at home that I could bundle under the same umbrella. I would, if truth be told, much rather be sitting in front of my computer optimising the code running my home server, or writing code for one of the sensors on my IoT network, or tinkering in my garage, or adding some new material to my book, or the myriad other things that I find more exciting and interesting.
All the current forms of AI that are being rammed down our throats (here’s pointing at you Microsoft) are great at creative endeavours; drawing, writing, researching, designing. Everything that I, and most other human beings, also love doing!
I do not want an AI to do these tasks for me so that it frees up my time for all the tasks I dislike. I would rather AI research looks into doing the menial tasks, freeing up my time for the things I love doing. Now that makes a lot more sense to me.
I wear two hats: one is firmly in the electronics field and here, I can marvel at the advances that have taken place with AI and especially inferencing at the edge, a truly useful adaptation of AI.
The other hat is in the programming/education camp. And here I have major concerns. Every student I have is heavily invested in using AI to produce work instead of creating code. I believe that coding is a creative endeavour, an art form if you wish. Trying to piecemeal AI-generated code together is never going to produce decent applications. After all, the AI model has been trained on information available on the internet – average code at best. Coding should be fun, and using AI is detracting from that sense of fun, excitement, and pride in getting something to work.
The companies that develop LLMs need to change direction and concentrate on developing AI that frees up our time, not so that we can have more time to do the tasks we don’t want to do in the first place, but rather to allow us more time to do what we love.
But I suppose monetising their systems is their first priority, not how they can BEST serve the human race.
Locate underground cable faults faster with the Fluke AF2082 A-Frame Comtest
Editor's Choice Test & Measurement
Underground cable faults can lead to costly downtime, service disruptions, and extensive excavation if the fault location is uncertain. The AF2082 simplifies this process by guiding operators directly to the fault point.
Read more...Here is what we learnt from three supply chain crises Seven Labs Technology
Editor's Choice
Component crises are not random events. They follow a pattern. After fifteen years and three major disruptions, Seven Labs knows exactly what that pattern demands, and what it takes to keep production lines moving.
Read more...The new reality of PCB material supply Jemstech
Editor's Choice Manufacturing / Production Technology, Hardware & Services
How resilient manufacturing strategies are helping OEMs navigate a changing global supply chain.
Read more...From the editor's desk: Local can be international Technews Publishing
Editor's Choice News
Welcome to the July 2026 issue of Dataweek. As you can see from this introduction, Dataweek’s regular editor, Peter Howells, is on extended leave and I am filling the void in his editor’s column – hopefully without being too boring.
Read more...Solving South African power problems with locally built intelligence
Editor's Choice Power Electronics / Power Management
Smart metering infrastructure in South Africa remains patchy. The rollout has been slow, coverage is inconsistent, and for the most part, granular per-unit measurement simply does not exist.
Read more...Lesley Havenga: Building partnerships for Africa’s electronics future
Editor's Choice News
As Würth Electronik expands its footprint across South Africa and the broader sub-Saharan region, Havenga’s blend of manufacturing expertise, supply chain knowledge, and people-centred leadership appears well suited to the task.
Read more...Generating negative voltages from a positive supply Altron Arrow
Editor's Choice Passive Components
It is common for IoT devices, industrial sensors, meters, and medical equipment to require both a positive and negative voltage, and this article explains the options available to produce a negative rail from a positive rail supply.
While every effort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily reflect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication. Advertisements and company contact details are published as provided by the advertiser. Technews Publishing (Pty) Ltd cannot be held responsible for the accuracy or veracity of supplied material.