How safe it is to rely on Chat-GPT in critical decisions ?

With the advent of conversational AI models like Chat-GPT and the reputation/publicity it acquired in the last few months, there would be increasing tendency to rely on this attractive and easy-to-use tool in deep management and consultancy issues we face everyday. But how safe would this be, taking into account, that such input from an automated tool would influence critical decisions to be taken by top management, decision-makers, and business owners?

IMHO, such a practice could be very dangerous and highly-risk taking into account the importance and criticality of the issues being addressed and the decisions to be taken, and their impact on the entities where such decisions are to be applied.

There is a very fundamental point in such reservation, and this has to so with the very basic nature between man and machine. the increasing capabilities of the machine, and growing proportional reliance on machine outputs in human life.

Few simple facts to keep in mind while dealing with this issue: -

1. Machines are just tools, invented to help man and make life easier for him
2. Computers and IT are just tools, very fact tools, and very fast idiots as a matter of fact ... at least in their basic structure.
3. Regardless how smart they may look, and how amazing they may perform - thanks to state-of-art programming - these wonderful things still do what they are told to do explicitly or implicitly.
4. A striking phenomenon associates itself with such developments in the form of inversely proportional relationship between man capabilities (shrinking) with machine capabilities (increasing). This has to do with lazy/relaxed human mind over-trusting the machine and what it can do
5. In my own words, artificial intelligence (AI) within the machines is no match to natural stupidity (NS) within incapable and unqualified human individuals

As a millennial cross-paradigm professional over the past 40 years or so, I have seen and observed what many youngsters have not come across in their professional careers .

Let me quote few of these things for interested readers

1. The subject of artificial intelligence is not new. It has been tackled through computers and computer technology since the early 80's of last century. Remember LISP and PROLOG ?
2. The term "intelligent" and "smart" has been used and even abused to describe devices (and systems) with some programmable functions and performance. Have you heard of smart cities, smart buildings, smart watches and other gadgets, and even smart toilets?
3. Biological studies, experiments and inventions on human things (like neurons and DNA) and the possible use within "hybrid" bio machines all the way up to very advanced human-form robots that started invading our life?
4. Medical and surgical pioneering work fed by humanitarian needs dealing with disabilities and loss of critical sense like hearing (ears), seeing (vision), movement (artificial limbs)...etc.

In many aspects, I consider myself fortunate to live and witness this evolution and maturing technical journey which merged and integrated the 3C (computing, communication and control) model to reality, something that I have foreseen and predicated since the early 80's (thanks to my engineering background in control and systems engineering). Today, I even feel more fortunate to see the new paradigm of IR4.0 that brings numerous hot technologies together under its umbrella to serve economy, industry and countless business domains in general. I will definitely cover this topic in detail in the future.