AI in business - opportunity, risk or threat?

There are many ominous discussions, emotional warnings, enthusiastic promises, and dreaming reasonings about AI and its role in our life the near future.

The main fear is that AI will harm humanity somehow.

Why? Because we don’t know what might happen with it.

That is uncertainty, which creates a risk.

But how significant is that risk?

The current implementation of AI is a brilliant mediocrity.

AI can be trained to perform known operations and combine them, but it cannot invent something new that we are not already aware of, such as a completely new engine instead of an improved one. A person needs to define the task, and a person needs to decide whether to accept the result or not.

Can AI quickly, suddenly, and secretly become an independent, self-sufficient, and dangerous superpower for our civilisation? There are no concrete arguments for it, only fears and fantasies.

Should we think about it? Yes, it is exciting.

Should we monitor that risk? Yes.

Is there a threat?

Yes.

It comes from an assumption that we are unaware of vulnerabilities and weaknesses in our enterprises. Vulnerabilities in cyber and other security, suboptimality in designs and operational processes, weaknesses in products and business strategy.

However, that threat is not from AI but from humans who can use AI for malicious activity and outperform us with better performance or more innovative solutions and strategies. 

We know how to manage such threats and risks. Conceptually, this is a well-known competition between a sword and a shield. We can and must use AI to help us with our current risks, issues and problems and achieve better results. AI will give us similar opportunities to close gaps, identify and remediate issues, improve productivity, design better solutions and run more optimal strategies.

AI's real risk and vulnerability for business is that optimisation will require new skills, solutions and infrastructure, which most companies don’t have now.

The good news is that methods and solutions for managing this risk are the same as those for taking advantage of AI.

What is the opportunity, then?

We can make AI an excellent assistant in human activities if we prepare our operational processes for it and train AI to do routine jobs. We can make AI a co-operator in routinised processes, which will require more preparation, training and adjustment, but the game is worth the candle.

AI innovation has significant prospects and a probability of a giant bubble.

AI potential is similar to the potential of the Internet or even entire computers and information technologies. Before AI, all computer technologies were just a crutch for human labour and management, while AI is the first step to getting real help and cooperation in the division of labour.

The Global economy and most businesses could benefit from AI by reducing costs, improving productivity, and creating new products and services.

AI optimisation changes will be mandatory and implemented promptly because slow actors and companies will lose the competition and be pushed out of the market. The key here is to optimise business efficiently and on time for a positive ROI.

In most cases, a transformation for AI will require the support of external consultancies and service providers with new consulting and service paradigms. The new paradigm is not replacing people in operational processes but designing the operational processes for robots and AI.

Some businesses will offer AI-based products and services and make excellent success from AI opportunities similar to mining tools suppliers' success at the time of Gold Ruch.

All enterprises (i.e. all companies or organisations; please see my post about enterprise) involved in an AI transformation should remember the high probability of a bubble in the AI sector. There have already been two AI winters, and the third will likely occur.

The main question is whether our businesses and economy have the intelligence, time, and resources to realise the potential of AI and manage the bubble.