Ovi -
we cover every issue
Μονοπάτι της Εκεχειρίας  
Ovi Bookshop - Free Ebook
Tony Zuvela - Cartoons, Illustrations
Ovi Language
Books by Avgi Meleti
Stop violence against women
Murray Hunter: Opportunity, Strategy and Entrepreneurship
Stop human trafficking
BBC News :   - 
iBite :   - 
The limits of automation The limits of automation
by Joseph Gatt
2021-01-05 11:49:46
Print - Comment - Send to a Friend - More from this Author
DeliciousRedditFacebookDigg! StumbleUpon

Imagine this world. All bank transactions are done via ATM. All retail store transactions are done by consulting online catalogues, and by making online purchases and having the product delivered by drones. If the product doesn't fit, you fill out an online form, a drone comes pick it up and you can exchange it, or get a refund.

Factories are completely automatic. Robots extract the raw materials and agricultural products from automatic fields and farms, robots transform and package and store the products.

ai001_400At restaurants and bars, you press a button and get your automatically manufactured dish or drink, delivered straight to your table by a tunnel system that is connected to your table. You pay automatically.

If you want a special order, or want to remove certain ingredients from your product, an online form lists the ingredients and suggests which ones you would like to remove or which one you are willing to take out.

The police force is automatic. Robotic vehicles patrol the streets and intervene when needed. Lawyers are also automatic, and a database informs whether the defendant is guilty or innocent. Doctors are robots; they scan your body and know what to do with it.

Is this world possible?

First off, this world won't happen overnight.

Rule number 1 of automation: it's individual institutions, organizations and businesses who are gradually going to adopt automation through trial and error.

Automation is usually resorted to when money needs to be saved in cost-cutting operations. For example, in several countries, automatic subway or metro sales points replaced the booths where tickets used to be sold. The machines that sell metro tickets are cheaper to use and operate than having to pay full-time staff just to sell tickets.

I hear people yelling: but that's cruel! Someone just lost his job selling metro tickets. Well here's my answer. In the 1960s in the New York City, London or Paris metro or underground or subway, you used to have ticket salesmen AND ticket “punchers.” That is in the old days there were no automatic turnstiles, and it was an actual person getting paid to punch your subway ticket. Just to say that automation is a gradual, evolutionary process rather than a revolutionary process.

So you're going to have different businesses, organizations and institutions gradually adopt automation in one area or another, through testing, trial and error, and we'll get used to automation before we know it.

But automation will, in my opinion, never be complete, and probably won't threaten your job.

Here are some problems that automation can, and will encounter.

Weather factors: electronic devices don't endure the heat very well, nor do the endure extreme cold. So you're going to have ABC corporations use nothing but robots in its assembly line. But on a 45 degree Celsius day, or in a minus 30 degree Celsius day, ABC corporation is going to have to fix a lot of robots, or perhaps even replace them from scratch. Too much electricity use can even make the assembly line blow up, especially in extreme heat waves or cold waves.  

Engineering factors: when you use nothing but robots on the assembly line, they are going to require a lot of fixing. You don't want angry clients at that robotic restaurant having to wait for 2 hours because the tunnel that is supposed to deliver their meal is clogged or just broke down.

So the mastery of the engineering aspects are going to take at least 10-15, perhaps 20 years, for any automatic assembly line, that is.  

Accidents: what if a robot kills an engineer or a client. Who's to blame? If too many clients or engineers get electrocuted by the robots, that's going to be a lot of blood money or manslaughter charges against the CEO of the company or the person in charge of the assembly line.

So if you do a cost-benefit analysis, and that you add in the millions of dollars in legal fees versus hiring 100 people and paying most of them minimum wage, you're probably better off hiring 100 people and paying them minimum wage.  

Engineer shortage: what if there are so many machines that the number of trained engineers doesn't follow? That means engineers are going to be scarce, and you're going to need to wait for days, perhaps weeks before you find an engineer who can fix the machine. Not to mention all those sham engineers who will come pretending to know how to fix your machine but actually aggravate the problem.

Dependence on automation: Let's say you set up an assembly line and it's all robots and nothing else. Let's say an entire nation completely depends on robots. What if there's a hurricane or a huge earthquake or a massive storm or flood and that the entire assembly line goes bust. That's going to be a lot of repair and fixing! And you're going to have to rebuild the automatic assembly lines from scratch.  

Automation as assistance: Truth is, most automation starts off as tools to assist individuals with tasks. In the old days you would bake the bread in a furnace, heated by wood. Then it was coal. Then It was electric furnaces. Then you had machines invented to knead the dough. Then it was machines to slice the bread. Soon enough it will be drones to deliver the bread.

Robots cost a lot of money!: There's this other problem. Let's say you use delivery drones. A drone costs something like 100 dollars a day to operate and maintain, give or take a few dollars. Plus you need an actual human being to pilot the drone. But let's suppose we invent drones with a GPS system and a drone that automatically knows what bread to pick up and where to deliver it.

How much are you going to charge for the delivery? Are you going to deliver that 1 dollar loaf of bread, bearing in mind that the drone could end up in an accident, could break down, could accidentally crash into a building and so on? Or are you going to demand that people order a minimum of 20 dollars worth of shopping? Or a minimum of 100 dollars worth of shopping?  

Training the users of automation: A lot of engineers have confessed to me that they are starting to grow sick and tired of new software and machines. Each time there's a new machine, it takes days, sometimes weeks to figure out how to use it. And by the time you've figured out how to use the machine, an even better one gets invented.

But the engineers are the backbone of any automated society. And if engineers massively quit their jobs to coach soccer or open restaurants instead, we have a problem.

How an automation economy would work: This is to conclude. In any automated society, the tools are going to gradually be invented to help individuals complete tasks.

I don't think we'll ever sit there and watch robots in action doing all the work.

There are certain things robots just can't do. Robots can't set consumer trends. Robots can't use common sense to assess real security threats. Robots can't see danger. Robots can't predict that a robot is going to collapse. Robots can't really take care of the marketing of a product. Robots can't fix themselves if they break down. And robots can't ask people real questions that demand complicated answers, among so many other things.

So if you're young and you're reading this, I would bet that by the time we live in a fully automated society, you and I will probably be grandparents, perhaps even great-grand-parents.

Print - Comment - Send to a Friend - More from this Author

Get it off your chest
 (comments policy)

© Copyright CHAMELEON PROJECT Tmi 2005-2008  -  Sitemap  -  Add to favourites  -  Link to Ovi
Privacy Policy  -  Contact  -  RSS Feeds  -  Search  -  Submissions  -  Subscribe  -  About Ovi