April 15, 2023


Will we lose their jobs because of AI or will machines wipe us out first?

Bored in the office? Here’s Tokyo police using drones with nets to catch other drones used by drug smugglers.

We haven’t had this much fun in years

Tokyo Metropolitan Police Department

Now, instead of manual labor, their work is much more like a computer game, and robots get you more impressive results than beat cops, in less time. With driverless cars now on the road, and usage of satellite data to effortlessly find illegal fishermen, it looks like someone’s job is getting easier as well as more fun. But while in this case drones and cops are on the same side, what about the rest of us? 

Is AI actually a physical threat? 

In the US in 2021 alone, the government is estimated to have invested $6 BN in autonomous AI-based weapons. How do they work? 

Without the need for human involvement, autonomous weapons choose and engage targets. They usually include armed quadcopters that can hunt for and eliminate persons who fulfill specific predefined criteria. Autonomous defense systems including weapons, smart vehicles, armed drones, and robots play a vital role in defense operations. These are fully autonomous tools that employ telecommunication technology such as the 5G network. Global military spending on autonomous weapons and AI is anticipated to grow significantly over the years.

Allied Market Research

Now, that sounds doubly inconvenient. You thought that the perspective of robots taking our jobs was bad enough, and now there’s autonomous AI that humankind’s best minds think we’ll be unable to control! That’s just great.

However, killing machines are reserved for criminals and mercenaries, so you'll be fine if you don’t break the law and stay away from war zones. Probably.

Is my job safe? 

According to studies, about 14% of people stated their jobs were already replaced by machines. As AI’s pace of development picks up, today T-1000 AI-based robots are becoming more effective at doing our jobs, and at a rather worrying pace. Adding one robot has the effect of replacing 3.3 people, according to MIT News

Think your job isn’t at risk? Here’s a small fraction of what Skynet was up to in the last few years:

– There is no need for accountants, since blockchain allows you to do away with paperwork completely

– AI is capable of processing millions of calculations in less than a second, while humans typically struggle to multiply 3-digit numbers, so trading has been dominated by AI for years now.

– Robots are much better at putting cars together than people. Just like truck drivers, warehouse workers make mistakes, sleep, and demand a salary. For employers, it’s nothing personal. There’s simply no comparison between human and robot workers in terms of effectiveness and costs.

– AI makes outstanding music (you guessed it, for free and in fractions of a second).

– Grammarly has already automated 90% of work proofreaders need to do, and the remaining 10% is soon to follow.

– Will we need programmers soon? As coders painstakingly one-finger punch the keyboard and take a while to think up code, AI writes pages of it in less than a minute. AI is actually already trying to cut developers out of the design process because it’s programming itself quite well. Now, imagine how fast programming will happen when AI is allowed to develop itself.

– Midjourney creates stunning images in seconds, for free, instead of artists taking months, not meeting deadlines, charging you thousands, and taking up time in administering costs. It’s not an exaggeration that AI will only take a few years to get better than humans at creating music, art, and literature based on the existing progress curve.

Using sound logic and facts to answer the question

At this point, you probably have quite a few objections. Are robots replacing humans? Preposterous! And you may very well be right.

What’s a good way to check the legitimacy of your objections and our theory? Think back and think how similar you sound to horse cart owners before horse carts were replaced by cars, Blockbuster before it went bust, and fans of paper letters before the era of emails.

Now, it’s not necessarily true that AI will outsmart and outnumber us all, but how do you prove or disprove that idea? With logic and expert opinions how.

Now, we’ve already addressed the logic component by listing above the professions AI and robots are better at than humans and supposing many others are to follow judging by the existing trend. Let’s move on to the opinions voiced by the brightest minds on the planet.

Here’s what Elon Musk thinks:

Should we develop nonhuman minds that might eventually outnumber, outsmart, obsolete and replace us? Recent months have seen AI labs locked in an out-of-control race to develop and deploy ever more powerful digital minds that no one – not even their creators – can understand, predict, or reliably control.

Pause Giant AI Experiments: An Open Letter by Elon Musk, Steve Wozniak, John J Hopfield, and hundreds more. 

AI being used to assist people sounds terrific, especially considering that that was one of the reasons why it was designed: doing repetitive, difficult, and dangerous work, and working in places where people just can’t go. But there’s a darker side to AI developments.

For example, this robot dog with a sniper rifle has a few tricks up its sleeve. Even AI itself admits that it’s a dangerous idea. 

Ghost Robotics, a military robot maker, has developed a quadrupedal robot dog that carries a SWORD Defense Systems Special Purpose Unmanned Rifle (SPUR). The SPUR is a high-precision killing machine that can hit targets 1.2 km away and is equipped with a Teledyne FLIR Boson thermal camera with a 30X optical zoom. The use of AI-powered robots with sniper rifles is not a job that should be assigned to robots, as it could lead to unintended consequences and harm to humans.


Does that sound a little too close for comfort? Maybe SPURs aren’t digging into your side just yet, but the government is definitely investing in this industry, and not just the government. AI has been developing at a rather staggering speed lately, and here are a few graphs to prove it, courtesy of our friends at Perplexity: 

– A graph from MLPerf that shows how faster training leads to better AI

– A chart from Brownstone Research that shows how processing power required to train AI on large data sets has doubled about every two years from 1952 to 2012

– A graph from Stanford University's AI Index 2019 annual report that shows how the speed of AI is outpacing Moore's Law

– Several charts from Technology Review that show the amount of money being invested in AI startups, the number of AI startups, and the growth of AI-related patents

– Various statistics and graphs from Oberlo that show the growth of AI in different industries

Final thoughts

Even though the rise of AI is every bit as impressive as we thought it would be, according to most research out there right now, like this study by McKinsey, only about 30% will be completely handed over to AI between 2016 and 2030. There’s no telling what the future holds, but it’s far more likely AI will make our jobs much easier and far less time-consuming like emails, taxis, and cars already did. 

About ROKY

ROKY is a digital production studio. We conceptualize, design and code all kinds of projects for brands, startups, and ourselves. We know that historically inevitably the only companies who survived were the ones that embraced new technologies first. That’s why we’ll be happy to help you integrate AI into your company’s processes in order to keep your employees, make their job easier and more fun, and increase their productivity exponentially. Get in touch with us!

  • Author

    Niki Kosmin

  • Share

Similar articles

How to survive the crisis and increase the effectiveness of employee training
Minigames from the New York Times. Why do they need the edition and how they were created
How does the Curiosity Rover operate