The Future of Work: Anticipating Jobs Yet to Be Created

Predictions are hard, especially about the future! Niels Bohr.

The above quote, often attributed to the famous Danish Physicist, reminds us of the challenges involved in forecasting. We live in a world of constant change, and the only thing we can safely predict is that the world will change.

Different types of Artificial Intelligence have significantly disrupted our world. Now, whether you are a techno-skeptic (constantly wondering whether all this technological innovation is of any benefit to humankind) or a techno-utopian (thinking that the world will be a better place with all the innovations), we are all thinking about the future of the world and how it impacts our future generations.

The jobs of the future that will be taken by our kids don't exist today, which is why it is critical to think about what skills to learn to stay relevant. Due to the rise of AI, people think that prompt engineering would be one of the hottest jobs in the market, but as my former colleague, Bill Franks argues in this article, just like there is no job title called "Search Engineer," we can safely say prompt engineer won't be a thing.

In 1784, the First Industrial Revolution followed the introduction of manufacturing facilities powered by water or steam, like the steam train, the steam engine, and the first mechanical loom. In 1870, the opening of the first production line at a slaughterhouse in Cincinnati brought about the Second Industrial Revolution. The Third Industrial Revolution coincided with the advent of Information Technology in 1969, which generated lots of data, but we could have put that data to better use. The data generated in Industry 3.0 was considered exhaust data, a by-product of doing business. The Fourth Industrial Revolution is characterized by digitalization. This is where it is all about converting the Digital Exhaust into Digital Fuel.

This nostalgic flashback is necessary because jobs and industries were eliminated at each stage of this evolution. Before Industry 1.0, in 1440, Johannes Gutenberg invented the printing press and, in doing so, put a huge number of monks out of business whose job was to write books by hand. At each industrial revolution, some jobs and industries were eliminated in favor of better productivity and returns. However, the difference is that change happened over multiple decades, and now the changes are happening over months.

AI has been winning art contests to build the world's first AI-based Software Engineer by Cognition .

The digital world has seen the benefits of Moore's Law over the last 60 years, so the problems considered hard a few years ago are now becoming solvable. This growth in Moore's Law is continuing, and Intel Corporation has promised to release a device with 1 Trillion transistors by the year 2030 (for comparison, an iPhone 14 has 15 Billion transistors). From now on, we will see a significant shift in each incremental capability as we enter the second half of the chessboard in Moore's law technology cycle.

Be comfortable with change

A few years back, I was asked to present to a group of graduating students. Their question was what field we should choose. My suggestion to them was:

"Observe in the last 10-15 years the jobs and skills made redundant by modern technologies. Stay out of those areas. Based on what you know today, predict which areas will be eliminated in the next 10-15 years. Stay out of those fields, too. Then, pick a domain; regardless of what you will pick, you will be wrong 10-15 years later. Because the technological innovation rate will be much higher than what your brain can imagine today, your imagination is limited by what you can tangibly feel and observe."

The key thing is working on the premise that whatever you are doing can be replaced by AI. We must prepare to make ourselves redundant every year. We must focus on transferable skills of different natures. In other words, "Jack of all trades has a better chance of survival than a master of one". What happens if you are a Master of One and your domain gets eliminated by AI?

What can we do?

A few years ago, one of my bosses provided me with a valuable framework for contemplating job automation and the application of human skills. Data is necessary for every task we encounter; the percentage required varies from person to person, ranging from 0% to 100%. If you adhere strictly to perfectionism, insisting on making decisions only when close to 100% of the data is available, you risk consuming excessive time and incurring unnecessary costs for the business by acquiring additional data. Moreover, if your decision-making process demands nearly 100% data, your role is at risk of replacement by a robot.

Conversely, you expose yourself to significant risks if you plunge into tasks with zero data. You can make a decision if 50-60% of the data is available and then supplement it with your human judgment (experience) to arrive at an informed decision. Should your decision prove incorrect, but you've utilized the available data judiciously and prioritized efficiency, a good leader should always back you.

System 1 vs System 2 Thinking

In his book, Thinking Fast and Slow by Daniel Kahneman; the author's main thesis is a differentiation between two modes of thought: "System 1" is fast, instinctive, and emotional; "System 2" is slower, more deliberative, and more logical. AI will start taking over a lot of System 1 type of work while the deliberate human approach to problem handling required by system 2 would still be a differentiator. While technology may be democratized, the business domain thinking would be required to make correct decisions, and this is where I feel most of the future jobs will exist.

Conclusion

I strongly believe AI is for humans, and humans are not for AI. Ultimately, there is no definite answer on what the future will hold. We need to spend our efforts on reskilling ourselves. I understand this may come across from a point of privilege. For some, it may not be easy to reskill themselves, and the rate of change is far too quick than we have seen in the past industrial revolutions. This is why governments are trying to enforce regulations.

With all the advancements of AI, we will see technology taking over type 1 thinking work, while deliberate, reflective, and wider-implication thinking will still be left to humans' type 2 thinking.

What do you think?

If you like, please subscribe to the FAQ on Data newsletter and/or follow Fawad Qureshi on LinkedIn.