The Human - Machine Symbiotic Future

Are robots really taking over our lives, our jobs and our identities? Certainly there has been a long-held belief that if we let artificial intelligence become too widespread, it somehow makes us less human. We need to feel like we have a purpose, and that purpose becomes less clearly defined the more power we give to machines. This has led to a certain hostility and skepticism towards machine learning and AI.

But the fact is, AI and robotics are nothing new. Although we are seeing a lot in the news now about recent advances with the technology, it has actually been evolving for decades. And society is just about managing to hold itself together regardless.

A brief history

Back in 1949, British mathematician Alan Turing already recognised the potential for machines to work alongside humans. “I do not see why it (the machine) should not enter any one of the fields normally covered by the human intellect, and eventually compete on equal terms,” he told The London Times. The following year he developed the Turing Test, designed to judge whether a machine could be distinguished from a human when using text communication – something that has become incredibly relevant over the past couple of years.

Lack of computational power was the main obstacle to Turing and others working in the field at the time. Although AI research and development was underway, computers were incredibly expensive and couldn’t store or process enough information.

A big breakthrough came late in the 1980s when IBM’s chess-playing computer program Deep Thought started beating professional players. Using what was, at the time, immense computing power, the machine was able to look 10 or more moves ahead to search for the best strategy.

At the turn of the century, AI hit the toy market with products like the Furby and the Sony AIBO robotic dog able to “learn” human voice commands. Development then shifted to more useful products such as driverless cars and voice-activated personal assistants.

Now, automation and AI are progressing faster than ever before. It’s not necessarily that our algorithms are getting much smarter; more that other technologies such as cloud computing and big data analytics are making it possible for machines to process much larger volumes of information – and learn from it accordingly.

It’s interesting to note that so far, this partial cognitive replacement of humans seems most prevalent in two quite disparate settings: heavily structured environments like factories and warehouses (Amazon’s assembly lines now come to the human pickers); and completely unstructured environments like our homes (say hi to Alexa and Siri). In between, there’s a massive, untapped market for indoor robots that operate in commercial spaces.

I’m talking about businesses like hotels, hospitals, offices, retail stores, banks, schools, nursing homes, schools, malls, and museums. Perhaps the delay is just because they are biding their time until the R&D dollars from the big tech have delivered sufficient advances and cost savings to make new applications more affordable. Once the sensors, computing hardware, algorithms, AI, machine learning, etc. are at this level, there is huge potential for companies to save money by using machines to replace human cognitive labour.

Cognitive Labour

Choosing how to spend our time has been an age-old dilemma for humans: do something that’s profitable but mentally demanding (cognitive labour) or something undemanding but also unproductive (cognitive leisure).

This may not be a decision we have to face for much longer, since robots now seem to be encroaching on what we thought was purely the domain of human ingenuity. Let’s look at the following examples.

Case 1

Can your robots play music? Nigel Stanford has worked with Kuka to develop one of the most engaging music videos ever. His video features robots playing instruments, and ultimately destroying them.

“Topics of AI, the singularity, robots and automation are always on people’s minds these days – and AUTOMATICA is the musical expression of these conversations. Really, the question is ‘What does it mean to be human in the digital age?’ Also, watching a robot make a piano explode is pretty cool,” says Stanford.

His work beautifully illustrates the fact that although robots are able to do amazing things, we still need amazing humans working with them to come up with the ideas and put everything in place.

As for robots composing music themselves, well, it’s happening, but you probably won’t be rushing to add these tunes to your playlist just yet.

Case 2

If there’s one industry people love to be wary of, it’s banking and finance. So when a solution comes around that bypasses human advisors and instead dishes out impartial or imperfect information on markets past and present, of course it will make a splash.

Robo-advisors are online, automated portfolio management services which cost next to nothing compared to a human financial advisor. They use computer algorithms to choose appropriate investments based on your risk tolerance and time horizon, with daily portfolio rebalancing opportunities when an allocation strays beyond a certain pre-set amount.

Startups such as Wealthfront, Personal Capital, and Betterment have launched robo-advisors as industry disruptors, while incumbents such as Schwab (Intelligent Advisory), Vanguard (Personal Advisor Services), Morgan Stanley and BlackRock have joined the fray with their own hybrid machine/advisor solutions.

Case 3

Returning to the idea of machines beating humans at games, Google’s AlphaGo made headlines last year by beating human players in the Chinese game of Go – which is far more complicated than chess. Not only that, but the program taught itself to master the game in just three days.

“The most important idea in AlphaGo Zero is that learns completely tabula rasa, that means it starts completely from a blank slate, and figures out for itself only from self-play and without any human knowledge,” said lead researcher David Silver.

The program, developed in London, marks a big step forward for machine learning. But the developers are clear that it’s a partnership; humans still had to design the computer in the first place. The team are hopeful that the technology can be used for wider applications such as medical research.

AI vs. Machine learning

It’s important at this point to draw a distinction between artificial intelligence (AI) and machine learning.

The term AI has been around for decades and it refers to the ability of a machine to perform tasks that mimic human intelligence. Practical examples include things like voice recognition, problem solving, and being able to hold a conversation.

AI can be general or narrow. Machines with narrow AI can perform one or two tasks with human-level intelligence, but are limited to these functions only. General AI, on the other hand, shows all the characteristics of human intelligence.

Machine learning is a particular application of AI whereby the machine teaches itself rather than requiring extensive coding and programming to be able to carry out a task. We saw a simple but clear example of this in case  3 above where a machine taught itself the rules of a board game – and went on to win against a human. Machine learning usually involves feeding a large amount of data into a program and letting it work out what it can learn from that information.

Your recommendation lists on Netflix, Spotify and YouTube are all the product of machine learning. These programs are constantly refining themselves by analysing immense databases of other users’ actions and preferences – and applying what they learn to your account.

Robot Density

While AI helps you discover great new music, robots are being put to more practical uses all over the world (including to produce what device you listen to your music on).

The automation of production is accelerating around the world according to the International Federation of Robotics: 74 robot units per 10,000 employees is the new average global robot density in manufacturing industries (up from 66 units in 2015). By region, the average robot density is 99 units in Europe, 84 units in the Americas, and 63 units in Asia.

The growth of robot density in China has proved to be the most dynamic in the world. Robot installations in particular helped drive the density rate from 25 units in 2013 to 68 units in 2016. Today, China’s robot density ranks 23rd worldwide.

Worldwide, the Republic of Korea has by far the highest robot density in the manufacturing industry – a position the country has held since 2010, mainly due to continued high volumes of robot installations in the electronics and automotive industries.

Singapore follows in second place, recording a rate of 488 robots per 10,000 employees in 2016. About 90% of these are installed in the electronics industry. Meanwhile, the Monetary Authority of Singapore (MAS) announced on 7 May 2018 that its S$27 million Artificial Intelligence and Data Analytics (AIDA) Grant has garnered strong interest from the industry. The overall aim is to foster a thriving AI ecosystem comprising financial institutions, research institutions, and AI solution providers.

Germany and Japan ranked third and fourth in the world, with 309 and 303 robots, respectively, installed per 10,000 employees in the manufacturing industry. However Japan tops the charts for global robot manufacturing: The production capacity of Japanese suppliers reached 153,000 units in 2016 – the highest level ever recorded.

Looking to the future

AI in the workplace

One of the main concerns about AI is that “the robots will steal all our jobs” – and with statistics like those above this doesn’t seem entirely irrational.

According to a McKinsey report “currently demonstrated technologies could automate 45 percent of the activities people are paid to perform and that about 60 percent of all occupations could see 30 percent or more of their constituent activities automated.”

Across the US, over 4.7 million people are employed in the computing sector as developers, software engineers, systems managers and teachers. Together, they fuel an app ecosystem worth $950 billion that has revolutionized the way businesses and industries operate, according to a new report from ACT, The App Association.

Earlier this year Ofo (a bike sharing company), China Telecom, and Huawei worked together on developing a narrowband IoT (NB-IoT) smart lock using Huawei's OpenLab, with the co-founder encouraging further collaboration on AI and IoT technology.

Ofo, now processes around 32 million orders per day, he said, with 10 million bikes, 200 million users, and 370 orders processed per second across 17 countries. The company has reduced global carbon emissions by 2.2 million tons, he added.

Smarter-than-us phones

Smartphones will be an extension of the user, capable of recognizing them and predicting their next move. They will understand who you are, what you want, when you want it, how you want it done and execute tasks upon your authority; tracking you throughout the day to learn, plan and solve problems for you. For example, in the connected home, it could order a vacuum bot to clean when the house is empty, or turn a rice cooker on 20 minutes before you arrive.

Today, smartphones that use machine learning depend on cloud servers, which limits how information is processed. Having a built-in artificial intelligence processing unit, however, could change that and increase a device's computing power.

British chip design firm ARM, the company behind virtually every chip in today’s smartphones, now wants to put the power of AI into every mobile device. Currently, devices that run AI algorithms depend on servers in the cloud, limiting performance by online connectivity which affects how information is sent back and forth. Few devices, run their own AI software. That might soon change: thanks to a processor dedicated to machine learning for mobile phones and other smart-home devices, AI smartphones could one day be standard.

Project Trillium would make this process much more efficient. Their built-in AI chip would allow devices to continue running machine learning algorithms even when offline. This reduces data traffic and speeds up processing, while also saving power.

With the advantages machine learning brings to mobile devices, it’s hard not to see this as the future of mobile computing. ARM, however, isn’t exactly the first in trying to make this happen. Apple has already designed and built a “neural engine” as part of the iPhone X’s main chipset, to handle the phone’s artificial neural networks for images and speech processing.

Google’s own chipset, for their Pixel 2 smartphone, does something similar. And now, their Assistant is incredibly human like! No need for actual irate, inefficient assistance from humans. Way to go Google Duplex Chatbot.

The issue of ethics

The future is half-human-half-machine. Elon Musk indicated that AI will be the best or worst thing ever for humanity. How do we ensure AI systems can be used responsibly?

If an AI system makes a mistake, who is responsible? As much as you can use AI and hear of its benefits, by definition AI will still bring about mistakes, some of which may be quite significant, even lethal, and involve multiple parties.

Many of our ethical priorities also depend on our communities. AI ethics must be context-specific with some values and non-negotiable with others in order for a machine to function like a human.

The Final Word

AI is far from perfect, given it relies on machine learning algorithms — still in early stages of learning — to understand tasks and achieve optimum decision-making. But the technology will improve, and businesses will absolutely need to ensure they balance the benefits of lower costs, increased efficiencies and innovation with the ethical responsibilities around liability and managing the impact of a technology that will disrupt jobs on a large scale.

Humans are creative by nature and I believe that we will continue to leverage automation, artificial intelligence and machine learning to augment the human skills and capabilities. In the future we will likely see superhumans charged with nano chip implants and machines capable of matching human emotional intelligence. So the success factor will be how to combine these strengths without losing anything in between. Aside from the huge business implications for every single industry, just imagine this future that will leapfrog humankind way ahead of the curve in areas like health and longevity, transportation and mobility, social inclusion and communication and even the planet sustainability and space exploration. Very exciting!

Innovation lies at the intersection of cutting edge technology and our life’s aspirations so we need to keep working on shaping a better future for everyone.