Why Should Androids Look Like Us? : an article by: GF Willmetts

February 4, 2018 | By | Reply More

‘Tortoise? What’s that?’

Although our current robots with latex skin are improving with every generation, people are still unsettled by them. Something in us recognises the nuisances in other humans isn’t quite there…yet. Personally, I think it has something to do with the eyes. Not enough muscles around them yet. Humans read eye tension a lot.

The reason to create robots and androids in the humanoid form is to prevent creating new tools and equipment when they can share our own. Why re-invent the wheel or, in this case, the tool? The humanoid-shape is compact, flexible, able to do large and small work and mobile and work in the same space as ourselves. Some small tools could be incorporated into fingers but then, no matter the grips, the functions remain the same. They can also be built stronger and faster. It’s also seen that having its appearance look closer to human would make it more acceptable but is that a good idea?

If this robot species looks too much like us, then how do we tell them apart? This is not to stop prejudice or assaulting them but merely to indicate that they are a separate species that lives along side us, not something that has to totally blend in. Our minds are comfortable with knowing what we have around us. Having the opposite for a species that is obviously stronger and smarter should mean having something that readily tells them apart or you’re likely to get paranoid about their activities. Other than the milky substance used for blood in the androids in the ‘Alien’ reality, you couldn’t tell them from human.

The replicants from the ‘Blade Runner’ reality even more so as under the skin they aren’t mechanical but synthetic organics. They might be stronger but they bleed red blood. In ‘Blade Runner 2049’, only an ultra-violent scan of the eyeball would reveal their serial number. Do we really want that as the only test to tell them apart? Even the original Voigt-Kampff test to look for emotions wasn’t much better. If you want to avoid the rebel replicant situation there, then it makes sense to make them look less than more human or at least different enough that they can be easily spotted. Bring able to tell them apart shouldn’t make them much different than any of the other colours humans come in, although I’d probably choose something a little different than colour although not necessarily like the fingers in the original ‘Westworld’ (1973) film.

The fact that they are stronger than baseline humans should deter people from wanting to destroy them. Of course, there is bound to be some sort of prejudice, humans are built that way, but androids would be used in places where human life would be at risk which would reduce human contact. Having them living amongst the normal human population shouldn’t necessarily happen. But then, why should it? Space in their bodies for a digestive system wouldn’t be necessary so they wouldn’t need to eat with us. They would be powered by battery, as with the androids in the various ‘Humans’ series or possibly a small nuclear reactor as with the covert Terminator models from their reality.

Humans have an innate desire to be friendly with anything from animals to computer hardware and software. How many of you here talk to your computer even before the current software that can obey verbal commands? How many of you have adapted quickly to using it? The same would undoubtedly apply to androids. How they look becomes less of an issue. If anything, it would help if they look less human so we don’t become too attached to them. Maybe then we would be less disturbed by their appearance or get too attached to not know the difference. Although I doubt if such robots would see themselves as only slaves and servants, total acceptance and not caring would get in the way of human development. We could quickly become our out couch potato society and not even blink. Leave the dangerous stuff like cooking food and cleaning to robots.

This doesn’t mean they have to look robotic but even robots can look friendly. You only have to witness Robbie from ‘Forbidden Planet’, the B-9 environmental robot from ‘Lost In Space’, C-3P0 in ‘Star Wars’ or even Sherman the robot from the 1989 film ‘Millennium’ to realise appearances can be ignored. Even when they look closer to human, like Lieutenant Commander Data from ‘Star Trek: The Next Generation’, they are acceptable. Data and the various other androids in all the ‘Star Trek’ series are no different to the other humanoid alien species that Starfleet encounter. Comparing Data to the earlier model Juliana Tainer, who is not even aware she is an android, it is invariably actions that are revealing than appearance. Some like Questor from the 1974 film ‘The Questor Tapes’ and Andrew from the 1999 film ‘The Bicentennial Man’ can even choose their own likenesses.

If anything, having androids who look too human also leaves the possibility of replacements as in the films ‘The Stepford Wives’ (1975, 1980, 2004) and ‘Futureworld’ (1976) if they are controlled by a human source. This might not even be an android’s choice but those who make them. Alternatively, as with the film ‘Surrogates’ (2009), just providing an alternative physical body for human hosts. Strictly speaking, there aren’t really androids because they don’t have their own minds.

The robot or android body is really only a vessel for a computer or artificial intelligence. As such, it is dependent on the protocols adherent to it. Initially and for extended periods, it will be on a learning curve to understand its environment and the people that inhabits it. With some of the latter, it will endeavour to copy and blend in, although I doubt, unlike the robots from ‘A.I.: Artificial Intelligence’ (2001) or the ‘Humans’ TV series (2015-), that it will be fully functional sexually. Sex would be meaningless as a means for procreation. How would an AI understand or enjoy sexual pleasure? It has hard enough time understanding the conflicting emotions humans have. Having its own set of emotions would cause enough internal conflicts, especially if they replicated things like depression, paranoia and psychopathic behaviour.

Other than being humanoid, anything else that resembles what we would describe as ‘human’ would be incidental to function. With processor speed, what would we regarded as slow robotic movement would probably not exist or just the limits of the body. When you consider how the new lighter bionic limbs are beginning to resemble normal function, limited only by power source. For a robot, this can be incorporated into the torso. With the new lighter lithium batteries, weight might no longer be a problem. An inductor field in the ground would provide unlimited power.

Modifying such androids to look less human doesn’t mean people can’t show affection to them, just an acceptance that they don’t belong to the same species. With AI, the feeling would also be mutual and would certainly be programmed where to draw the line. Whether this would ensure them to be slaves or servants or never have a desire to revolt against their masters would be hard to conceive. The limitations come down to the programming. Without an emotional make-up or one compatible to human, such things would be beyond any such android. It would be like trying to understand a geyser of water without an understanding of pressure and heat. Even if it were possible, there is just as much likelihood that the android would fake such emotions than to really want them.

The desire for robots or androids to be more like us is a human desire and unlikely to be that of the machines themselves. The only thing they would really need is the access to knowledge. Equal AIs would draw the same conclusions.

When you consider the violence humans inflict on other humans, they have as much chance of assault as any other humanoid with one difference, they have less chance of being damaged. Well, short of being squashed in a hydraulic press, but that depends entirely what the android chassis is made of.

Should it be possible for Asimov’s Laws Of Robotics be programmed into an android, it should be capable of protecting itself from harm providing it doesn’t kill the human(s) attacking it. There are multiple ways of disarming, stunning or rendering any human unconscious without physical harm. Unconscious is a good option as it prevents further injury. Whether a human dies by getting in the way of another human attacking breaks such a directive is an interesting thought but it is not in a direct action it is taking.

We have a wide selection of Science Fiction stories that clearly shows the pitfalls of making robots and androids resemble humans too closely. It’s a lesson we should pay heed to in terms of making them clearly identifiable than having to run a test where even some humans could fail.

© GF Willmetts 2018

All rights reserved


Category: Science

About the Author ()

Geoff Willmetts has been editor at SFCrowsnest for some 21 plus years now, showing a versatility and knowledge in not only Science Fiction, but also the sciences and arts, all of which has been displayed here through editorials, reviews, articles and stories. With the latter, he has been running a short story series under the title of ‘Psi-Kicks’ If you want to contribute to SFCrowsnest, read the guidelines and show him what you can do. If it isn’t usable, he spends as much time telling you what the problems is as he would with material he accepts. This is largely how he got called an Uncle, as in Dutch Uncle. He’s not actually Dutch but hails from the west country in the UK.

Leave a Reply