The Seduction of Automation

Automation seduces with its promise of error-free operation and labor-saving efficiencies.

In my case, the goal is to be on a fishing boat (or out doing RF surveying) while the network is looking after itself and only be notified if there is a problem that can’t handle itself. 😉

AI promises to fulfill this goal. Maybe. But that topic is for another day.

Automated systems are trusted to perform complex tasks with precision and consistency.

However, this trust can lead to a phenomenon known as "automation complacency," where skilled individuals become overly reliant on automation, assuming that the software will handle any challenges that may arise.

If employees become too dependent on automated systems, they may lose their proficiency in certain tasks, creating a skills gap and hindering adaptability in the workforce.

To mitigate the risks associated with over-reliance on automation, it is crucial to maintain a balance between automated assistance and human judgment.

This involves designing human-centric AI systems, setting clear boundaries for system usage, and providing robust training to ensure users remain critical of automated outputs.

However, when building such a system one thing consideration must be in the forefront. Who will fix the system if/when it breaks.

And this is the main "scary part" of AI. Whilst AI helps you design and build great automation systems and workflows, maintaining these systems is a different matter altogether.

No matter how skilled you are, unless you have an intimate knowledge of what the system does and more importantly how it does it fixing it can take a considerable amount of time.

I have seen a few automation systems that hide, or more nefariously, obscure the workings and as a result are unnecessarily complex and difficult to maintain. This is exacerbated when the system has been in operation for some time and anyone that knows about it has long left for greener pastures.

When designing an automation system, the main objective should be to make it simple. Simple to operate, simple to manage and simple to maintain. as far as I can see AI fails in the last regard.

Please comment with your thoughts on using AI for automation.