“The Mars Curiosity rover is equipped with “AI” technology that selects research targets.” This statement introduces Elizabeth Howell’s article in Seeker (published 08/22/2017). The article presents NASA’s current and planned future use of robots with “AI” on Mars and potentially for missions beyond the solar system. But how else will we use Artificial Intelligence (AI)? Where can AI take us? Where are we allowing AI in our lives? Is it taking us into the realm of science fiction and a world apocalypse as pictured in the Terminator movies? Or is it taking us into a highly promising realm in which information can be gathered, and decisions can be made rapidly resulting in an improved life for humanity?
For the Mars rover – or any other expensive and far-off piece of rolling stock, we do not want it to have to query us – its operators/mission-controllers – whenever it perceives an issue with its surroundings. We do not want it to have to ask us whether it should stop or turn around or go straight whenever it reaches an obstacle. Especially if it takes 30 minutes (Mars approximate query and response time) or 2 hours (Jupiter approximate query and response time). We can all understand that a piece of equipment or a scientific instrument which can operate independently may be a greater benefit to us and perhaps to the device itself.
Now consider something closer, the common household thermostat. Thermostats operate on their own once they are programmed to do so. If your thermostat had to ask you if the house was too hot at 82 degree (F), and you as the mission-controller had to get up and touch it or yell across the room to tell it what to do – you would not be satisfied. The system would not be efficient. There would only be the perception of the surroundings by the thermostat but no helpful outcome. The type of independence found in the common thermostat requires that the robot – or independent operational equipment (InEqu) – must progress beyond a perception of its surroundings and take an action.
The type of independence needed in the Mars rover is of a higher level. The rover needs to be able to operate with a high level of independence in order to fulfill its mission. It must be able to perceive its surroundings, determine possible alternative courses of action, analyze the potential outcome of an action, then make a decision, and act along the chosen course of action.
This decision-making has long engrossed science fiction fans and generated countless arguments concerning the application of Isaac Asimov’s three laws of robotics. These laws establish a progression of decision-making to protect humans as well as the robot itself. These ideas intrigue us even though we are barely on the threshold of AI. How can we be certain that a robot on Mars will follow a given process of decision making that will enable it to protect itself and thereby protect its mission?
Also, what are these robots that can take independent action? Please do not picture a bipedal, smiling ape visage. Yet that is what we most often think of when we hear the term “robot.” I would venture to say that the vast majority of robots around us today look nothing like a human or a rabbit or any creature that we may try to envision. A welding robot which looks like a disembodied ant eater is probably one of the more common robots. It has no need for legs or for a face or for friendly features. It has a task to do and a place to fit. I will certainly agree that when we humans have to deal directly with a robot it is nice to have a familiar appearance or a cute face so we are at ease in our dealings with them. I raise my hat to the robots in the Henn-na hotel in Nagasaki as reported by Monisha Rajesh in The Guardian (08/14/2015). A smiling lady – a velociraptor, both of which we all know and love. These androids (from the Greek as in human-like) are willing to interact with us due to their programming. We might be willing to interact with them as they do not appear any more dangerous that a kitten. Japan will likely continue to lead in AI/InEqu. Their preparations for the 2020 Olympics include AI in transportation, security, and in traveler’s assistance such as language translation and general information.
But back to the question – where are we taking – or being taken – by AI in our lives? The initial answer is we, the builders of the InEqu (pronounced as ‘any-que’), will allow AI to take us as far as we program it. But then there is the case of Facebook’s robots as presented in Forbes.com by Tony Bradley, 07/31/2017. These devices remarkably went beyond their programming. Here was a system created to support a Facebook process, but then the AI devices developed their own language to speed/enhance their ability to communicate with itself/each other. Isn’t enhancing speed and operability what we want? To be able to sort information quickly? To be able to improve processes quickly? In this case, it might not have been a total surprise to Facebook’s “people”, but to the rest of us it was a “Wow” moment. Hopefully one in which we said, “Wow – isn’t that cool,” rather than “Wow – shut it off and don’t do that anymore.”
Only two decades ago many of us thought that a vacuum cleaner that operated by itself and could scare the pants of a cat was pretty sharp. We have moved beyond that quickly and will continue to do so. This, like the thermostat, may be considered “little AI”. Yet – again – where will we end up? I believe we will not end – but will continue to progress and to harness the ability of an InEqu to analyze (think?) and act independently for our benefit. In other words, to help us pursue and uphold our inalienable rights.
My first interaction with “big AI” will likely be the self-driving car. No kitten this, and hopefully my first interaction will not involve insurance companies. But these large, metal moving machines are just the InEqu’s that we are driving towards (pun intended).
I love to drive; in an active sense. I get into my car, and hands on the wheel I drive to work. I may listen to music on the radio. I watch the other drivers and pay attention to the road. I see a driver over there paying attention to their text message. I see another in animated conversation with a person that may be on the other side of the planet. For those people (assuming they are not androids) self-driving cars cannot come soon enough. But for me I will stay in my big old Detroit steel shell and enjoy the road. It is my cocoon against the assault of AI. It’s coming sure enough. I’m just not ready to participate. This then is my declaration of being independent.
Wired magazine in their August 2017 issue asks if we fear the future. I say, No. I look forward to the future. I just hope that when the texting population gets to their destination that the AI in their Detroit steel is programmed to wake them up and ask them if they want to take their coffee with them. Wired magazine enjoys writing about AI. And we should be glad that they do. It keeps alive the idea of machines/robots/InEqu that can do their jobs/processes quickly and efficiently and allow their human counterparts time to improve the overall process. This is not an us versus them; it is just an us – InEqu and people. In the meantime, enjoy the science fiction and stay in touch with will be appearing at the 2020 Olympics in Tokyo.