Ok, we were on holiday and there was likely wine involved in the second picture. The conversation was likely different too. Today's was about AI systems in management and how it could affect decision making. Sound familiar? Sounds like something I might be talking about except, it wasn't, it was her, my wife talking about AI. You might be saying that it shouldn't be a big deal but, you don't know her.
She is planning her thesis which will be on communication in corporations and got the idea from somewhere that since Artificial Intelligence is going to start creeping into the arena at an increasing rate, it would be an interesting and relevant topic to do her masters thesis on. I agree, it is going to increasingly encroach upon our decision making and affect all kinds of positions we hold.
One thing is that an AI never need make perfect decisions, it only needs to make better decisions than us and this is quite easily achieved through data collection. If there are two employees doing the same tasks and their actions, decisions and habits are tracked and connected to future results, combining the two and doing what works and eliminating what doesn't immediately improves the process. Even though this might have other implications and result in new errors, they too will feed back in and adjust AI behaviour again.
Let's say there are 10 employees doing the same thing and one makes a mistake that needs correcting. Training needs to be developed that can correct it, it needs to overcome human resistance, deployed, tested and it takes time for it to sink in. If an AI needs to make a change, it makes it immediately and will not make the same mistake again. If this is applied to all mistakes, even though it might have a very steep learning curve to begin with, it will very quickly advance.
How does this affect management decision making in the future? What management? Even if jobs require people to perform them, management level is just a communication function that makes decisions and distributes the information across the network. In time, humans are going to be inefficient at most of this meaning that much of the middle management level positions become bloatware and unnecessary. There will likely still be strategy makers but they will be informed by the AI that makes far fewer calculation errors with much more fine-grained information than any average manager could.
Average is an important concept here because that is essentially what the AI is competing with in the early phases of adoption, the average. But, it doesn't have to learn from the average, it can model the best in the same way neuro-linguistic programming methods attempt to improve skills through modelling the best of the best. The difference is that the AI is able to adopt new practices immediately upon making the decision it needs to.
This changes the way companies operate at fundamental levels because companies themselves are designed to continually reach for efficiency of practice to maximize their returns for shareholders. They are not dependent nor attached to any individual within the organisation and take a all are expendable approach. This is the perfect environment for a base level AI as it doesn't have to consider the social and community implications of its decision making as long as it stays within the guidelines of whatever laws it is governed by.
What effects will this have on how communities and all dependents on corporations including governments function is yet to be seen but perhaps, it is not as dystopic as I think it could be. Maybe, we can utilize these tools to better manage base tasks and maximize our own opportunities to have space to do more of the things we enjoy in this world like, going to breakfast with our partners.
With automation, will more genuine smiles return?
Taraz
[ a Steem original ]
