Artificial Intelligence — What’s Worked So Far

It’s the third week of February 2018, and these are some of the tech headlines:

• US risks losing artificial intelligence arms race with China
• People freaked out after robot dogs opened a door
• Artificial intelligence: The saviour of mankind or the end of the world?
• Taryn Southern’s new album is produced entirely by AI

What’s the common thread? Artificial intelligence.

Countries are putting AI on the national agenda:

CEOs are eager to see AI mentioned in their company’s press releases:

Moreover, Google search trends for AI are skyrocketing.

Gartner Analysts predict, “By 2020, AI technologies will be virtually pervasive in almost every new software product and service.”

Artificial intelligence has arrived.

AI-powered technologies are everywhere. Whether that’s an inherently good development for our future or real reason to be concerned, even a little frightened… that remains a matter of perspective.

For now, let’s focus on the positive, the irrefutable benefits of artificial intelligence and what’s worked thus far.

Wait, what is AI exactly?

Simply put, artificial intelligence helps robots act like humans.

The three stages of AI are:

• Narrow (for specific tasks)
• General (applying knowledge from one domain to multiple other)
• Super (way smarter than us).

What AI can already do today

What makes humans unique is their ability to think and act. Take a process such as differentiating photos of cats and dogs. While this task may be simple for us, the required method is somewhat complicated for any other being or even a robot. This is where we have the edge over animals and machines.

With AI, we are trying to replicate different ways of humaneness. On that note, let’s look at some AI awesomeness that’s already happening:

Wow, that’s pretty advanced stuff, you think? Keep in mind that all these progressive technologies (still) fall under “narrow AI.” Meaning, you feed a machine with data and algorithms, and it excels at executing a specific task you taught it to do.

Imagine the next stage, when machines will be able to acquire expertise in one area and apply it to another. This will be when general AI (also called “strong AI”) has arrived.

General AI — the next stage of artificial intelligence

“General intelligence is what people do,” says Oren Etzioni, CEO of the Allen Institute for Artificial Intelligence, Seattle. “Things that are so hard for people, like playing championship-level Go and poker, have turned out to be relatively easy for the machines. Things that are easy for a person — like making sense of what they see in front of them, speaking in their mother tongue — the machines struggle with.”

General artificial intelligence can take lessons learned from previous efforts and apply them to solve new tasks. General AI will be able to take on an ambiguous function, such as “solve this problem,” and deduce the way to do it through required reasoning, experimentation, and understanding. Just like us, general AI can learn.

In short: General AI will be adaptive.

AI will make humans smarter

When you are driving, AI will tell you how to detour around heavy traffic. When you are eating, AI will tell you what and how much to eat to stay healthy. When you’re at the supermarket, AI will tell you that you can quickly find your favorite bread in aisle six, on the second shelf to your right. When you come home, AI will set your desired room temperature and dim the lights when it’s time to sleep.

Will AI take away our jobs?

Researchers believe that the automation apocalypse is already here. McKinsey recently studied what kind of jobs are easiest to automate. The verdict is that machines incline toward tactical applications.

Algorithms have already replaced staff managing medical or legal records, accountants, auditors, as well as financial and sports news reporters. Bank tellers, video store employees, travel agents and gas station attendants have been replaced by ATMs, Netflix, airline reservation systems and self-service gas pumps.

However, Oxford academics researched 702 common professions and found that jobs involving creativity, building complex relationships, intuitions and wisdom are future-proof.

So, with all the negativity building from people saying AI will eat our jobs, this study also found that there is a new host of job opportunities in our future, thanks to automation.

Repetitive tasks, such as the work of telemarketers or loan officers, would go away. New jobs, such as cryptocurrency regulators, drone traffic optimizers, and brain quants will expand.

Futurist Martin Ford in his book Rise of the Robots: Technology and the Threat of a Jobless Future explains: The jobs most at risk are those that “are on some level routine, repetitive and predictable.”

“Evidence shows us that if technology destroyed jobs, there would be no work today for anyone. The technological revolution we have seen in the past 30 years has been unparalleled and exponential, and there are more jobs, better salaries. The best example is the German region of Baviera, one of the parts of the world with a higher degree of technification and robotization, and with a 2.6% unemployment. An all-time low. The same can be said about South Korea, and the world in general.” — Daniel Lacalle, Economist.

We need to develop the soft skills to compete with and complement AI. We have done an excellent job at maintaining our dominance so far. I trust we will continue to do so.

Staying ahead of the AI curve

The idea is to smooth the transition to the age of artificial intelligence.

AI is now going mainstream: Algorithms and supercomputers, once limited to specialist researchers at government labs, are now accessible to startups and private companies. The demand for tech talent is growing.

Alan Stukalsky, chief digital officer at Randstad North America, says, ”Employee training should be considered a company investment, not a resource drain.”

“Investing in artificial intelligence (AI) and human-machine collaboration could boost business revenues by 38% by 2022 and raise employment levels by 10%.” — Accenture, 2018.

To be empowered by AI as organizations — instead of worrying the automation revolution will roll over us — we can:

• Hire the right people who have a learning mindset
• Recruit more big-data-science teams
• Encourage a culture of lifelong learning
• Focus on improving emotional intelligence and the ability to communicate
• Seek community support on platforms like AIforums and Github
• Add machine learning to your products, request APIs, train and optimize apps (AWS, Databricks, Google cloud, Azure, Sigopt, etc.)
• Upskill staff. Use online platforms like Coursera, Udacity, Udemi, Codeacademy, Big Data University, and Microsoft’s edX. Tech firms like Amazon have initiated programs such as Career Choice, which pays up to 95% of tuition fees for employees who want to learn new skills
• Follow the right news sources and influencers.

In short, anybody can have an edge on competitors with the help of AI by getting small build-ons, making awesome products, lowering costs and, thereby, delighting the end user.

It’s easier than you can imagine.

Jump in, lead the pack

For virtually any organization to be successful in the future, senior management first has to understand how automation can transform their company and then implement new business processes enabled by AI. It’s never too late to start. Better understanding intelligent automation can also help business leaders to identify which tasks within the organization can be done better by machines and how to free up human resources to focus on key areas algorithms can’t replace — at least not for a while.

Being a progressivist, I believe that AI would be a great learning opportunity and, if done right, can truly bring about positive change in the world. Just like how horse-drawn ploughs revolutionized farming in the 1100s and steam engines drove manufacturing and transportation into a new era in the 18th century, AI can be the key player in the next century of digital innovation.

An Australian study on automating the job of bank tellers found: “While ATMs took over a lot of the tasks these tellers were doing, it gave existing workers the opportunity to upskill and sell a wider range of financial services.”

We have a long way to go. But rather than worrying about AI taking away our jobs, let’s focus on improving our skill set and productivity. With the right approach, the benefits of AI will dwarf its tradeoffs.