“Don’t worry, human intelligence will never be replaced by machines.” That’s what I was told as a freshman foreign languages student at a university. That was the time the concerns about the machine translation taking over the human, first came up. For an honest average playgoer, language is nothing but a set of words put in a specific order based on some (not so) simple rules. Learning languages is a grind. Knowing languages is extremely rewarding.
It’s absolutely natural that we started dreaming of delegating the grind to the machine while enjoying the fruits of its labor as merited creators of the technology.
However, everyone who’s tried working with machine translation quickly got put back on their heels. Machines make imprints of languages based on the rules provided by humans. Along with that, sentiments and emotional contexts shape the core of languages and reflect the identity of the people speaking it.
AI takes over language UI
With that said, Artificial Intelligence is still a strong asset in our pursuit of increasing the speed of everything we do and ridding us from the grueling aspects of our lives. Learning languages became a much quicker process as compared to 20 years ago. It’s not because we got smarter or languages became simpler, it’s the media that made learning more accessible.
In other words, we received more UI controls of foreign languages, allowing us to retain the sentiments while passing the basic operations to a computer. For example, how much handwriting do you do these days? How much do you rely on spell checkers and keyboard auto-correct function? Do you dictate messages to digital personal assistants like Siri and Google Now? I mean do you even need to know handwriting these days in order to be a fully-functional member of the community?
Those are all the skills we once had but at some point, thought it would be cool to hand them over to the computer because those skills are repetitive and somewhat redundant. Well, it’s use it or lose it, isn’t it? For instance, calligraphy used to be widespread among a lot of people. My father’s handwriting was one of the most beautiful writing I’ve ever seen. The keyboard was introduced to speed up the process, then it became predictive, at some point, it will be intelligent. And next thing you know, you don’t have to provide any input at all other than the mental one.
Calligraphy used to be a school-taught skill. Now it’s an art form.
Where does delegation stop?
My question is if we keep improving technology to take the workload off ourselves, how many of our essential skills will become forms of art? Currently, we are willingly giving up manual and repetitive work with parts of the interfaces for a chance to skip what we’ve mastered.
Dropping the old ways is the essential requirement of technical advancement but where do we stop? Do we stop when we have to work 3 days a week rather than 5? Maybe one day? Or no work at all? If not, what types of skills do we reserve for human labor only? Massages and eating contests excluded. Some of you might say:
“Yeah I might have given up the form but the substance… a machine will never have my personality.”
True, an AI won’t tell you a dick joke. It won’t have a drink with you. Won’t piss you off with its inability to understand the simplest thing. These are some of the skills that make us us. But what if the machine can learn every single punchline of every single joke? What if its vernacular contains all the words, can it give you the best pun ever? That’s impossible… or is it? We already have services providing sentiment analysis. Machine learning technologies allow everything we put out to be recorded and dismantled, thus reproduced.
The chemical synapse often ends in a gain, which means that the postsynaptic neuron is stimulated the same and often more than the presynaptic neuron. Compared to chemical synapses, electrical synapses conduct nerve impulses faster, which means provided enough gain, electrical synapses running AIs can beat human brain’s non-linear performance along with the undeniable advantage in speed.
Sam Harris gave a wonderful projection of this idea in his TED talk. “Electronic circuits function about a million times faster than the biochemical ones, which means a machine can think a million times faster than the minds that built it. So you set it running for a week and it will perform 20 000 years of human-level intellectual work. Week, after week, after week.”
All of a sudden we don’t seem so immune anymore. We have feelings, but we are limited in our means of expressing them. AI is only limited by the hardware capabilities and once that ceases to be or reaches the point in time where the average hardware will be enough to outsmart average people in all aspects of life, machines will operate on our emotional level without being sensual at all.
Sentient is the new sensual.
But AI can’t design?
Design itself is not a skill that can be passed on or taught in one specific unhindered way. Design is a process of evolution and adaptation, affected by the environment as well as deeply subjective circumstances. The amount of experience a designer might have accumulated does not necessarily mean superiority in the product they create. Different social, business, and technical variables are wrapped into a creative form to define a product foundation. On that foundation, we create features which satisfy the needs of users and meet the requirements of owners.
The success of any UX design is the satisfaction of users. We use different data channels to understand users and speak to them in a better way. Everything we know comes from these data channel. The more information we accumulate, the bigger chance we have to fill the needs.
As 2018 has already shown, no data goes untouched on the internet. Everything we read, post, send, and save is being stored somewhere waiting for its chance to help someone do something ?
But enough with conspiracy, let’s talk UX. The personal data we willingly give away to our digital profiles are being picked up to provide better service for us in the first place. A more tailored experience is the goal of the most of personal data gathering techniques.
However, everything on the web can be faked. Catfishing came to exist as an exclusive online phenomenon because literally, machines will except everything you put out, including the blatant lie.
Our bullshit detector works way better in real personal communication. We rely on a physical response when we communicate which often happens subconsciously. 55% of a message is conveyed through nonverbal elements like facial expressions, gestures, and posture. Add to that the uniqueness of every human being, and you’ll get a scenario where physical information can’t be observed. At least with the tools we have today. However, it’s cool to think how someday AI can pick up the subtleties of our verbal and nonverbal input. This could possibly make the entirety of device controls redundant by shifting the focus from a direct input to the intuitive one.
The bad news is this will be the end of anonymity as we knew it, which in the world of savage capitalism is #highlylikely to be used by an authority of some sort.
The good news is, there is no sign of such an advance yet. Face recognition is not even at the level of a toddler and we’re talking shades of meaning recognition?
UX design is safe
The design job ain’t going nowhere folks. Even with a tremendous leap in the AI technology, the first jobs to go will be the ones involving the legwork. When and if we do create an intelligent machine we can trust like we do a person, the jobs of drivers, consultants, secretaries, support specialists, and anything that is based on the repetitive work and with a somewhat low learning curve. Technological advancement relentlessly tackles manual labor.
“Muscles aching to work, minds aching to create – this is man.”
― John Steinbeck, The Grapes of Wrath
Design being a creative job operates on the level beyond attainment of AI as we know it. We have the ability to influence people’s minds through a context we put things in and through our own empathic experience we rely on in creation.
Even the AI features we create reflect ourselves. We tend to personify technologies because we gravitate towards the things we know and understand more than to the unfamiliar and alien ones.
“Just because” and “why not” are still solid points in reasoning for a particular design move.
With that said, machines and designers are not in the relationship of antithesis. Designers are more than willing to incorporate smart software and applications into their workflow. In fact, they are the ones honestly believing in the power of human-AI synergy. At list in design.
Tame the beast instead of slaying it.
Here’s where design can benefit from AI:
Delegation of repetitive work
Even such an internal and emotional industry as design requires a certain deal of mechanical work. Every new and improved design tool out there offers a great amount of automation.
Let’s face it; digital design’s pet peeve is development …and testing. The fact that you are done designing a product doesn’t necessarily mean the product design is done. In fact, you will only be able to enjoy the fruits of your labor once a lot of other people have their say. At some point, there has to be a way to introduce new products without extended production stalemates.
Intuitive design systems
Designing in a systematic approach is incremental for web application production. This is where you have to keep your creativity in check, observe business logic and consistency. From all, I think this is where AI will make its first notable imprint in the industry. Going back to my language analogy, machines can learn patterns, variables, rules, and modules which constitute any design system. This means humans will be able to step down from the meticulousness and rely on robotic precision more than they do on human memory and attention.
Visual sentiment analysis
Essentially, every social media platform today has some sort of machine learning system set up to analyze user data and pitch related content. Either based on the activity or word match principles, apps know which product they suggest has a higher probability of selling and which content keeps users in the loop. What if visual design could do the same?
This would probably make it possible to categorize creative work, put it into virtual libraries, learn the dependencies and perhaps understand the deep-seated structure of design. Could this become the beginning of AI design based on the analysis of thousands-of-years-worth of human art and design evolution?
Why AI needs no UI
User interfaces appeared because there was no other way to communicate your intention to a machine. A mechanical input was the only reliable method of conveying your command. Literally, something had to get hit with a hammer for it to shift some other thing in the mechanism and forward your initial input to the parts that will perform a task for you together is a predefined order and with a predictable outcome.
Well, AI is here to dismantle this order. When we talk to another human and explain our intention, we don’t need to shake them by the shoulders or tap them in the forehead. Nevertheless, people pick up what we’re laying down, get the meaning, the intention, project the outcome, and form their attitude to this information.
AI doesn’t need UI. Smart interface is no interface. The battle will go down on a different technological, ethical, and sensual level. As of yet, we are nowhere near this. Neither the understanding how to make material things without using our kinetic energy, nor the technological resource to make it happen.
This is what AI UI is like. An abstract object moving in within Newton’s physics logic:
Only after we master the ability to materialize things from the air, utilize quantum computers to generate data, and operate machines with our minds, can we think of the real synergy with AI. Before that, our little robots will be our assistants and will do only what they’re supposed to do.