In 2013, Google created an AI program called Deepmind1 that managed to beat an arcade game in less than a day.
The programmers simply gave the AI program just one objective, without instructions on how to play the game or anything else: reach the highest score. And it took Deepmind just a small amount of time to “crack the code” and beat the game.
This shocked everyone! It was as if the program could think for itself and learn from its mistakes: for it was always said “to err is human”.
This seemed apocalyptic: the late Stephen Hawking famously came out to say that AI will bring the end of humanity.2
Robot + AI = Terminator?
The COVID19 pandemic has pushed the trend for online grocery shopping; and with a lot of workers stranded at home due to lockdown (and the infamous Pingdemic3), a lot of grocery warehouses have started replacing the human workers with robots.
I suggest watching the video below:
These robots use an AI software (see latest IPO of Autostore4) to calculate the fastest route to pick up the different items without bumping into one another.
A lot of other jobs can (and were) automated and replaced by robots or simply a computer program. Take data entry for example: in the last 30 years, companies used to hire people to simply enter data into computers. They also hired supervisors to make sure the data entered was correct and no errors were made. All these jobs were replaced.
But the fears of robots replacing the workforce can be sometimes exaggerated. Remember when people thought ATMs would replace bank tellers: in fact, the number of bank tellers increased.5
Robots (+AI) are actually essential and needed for economies to grow and increase production.
Not only that, in a country like Japan where you are having a large number of elders, robots are also playing their part in taking care of them: for example, AI is used for robotic care takers and for self-driving cars to drop them off at the nearest hospital.
Quis custodiet ipsos custodes?
One reason not to worry too much about AI is that it is only a code programmed by humans with set parameters to perform set tasks. So if a Jordanian company claims to have “AI” in their product/service, it is nothing fancy or glamorous, only some lines of IF/Then code.
What economists, and policy makers, on the other hand should be worried about is how to balance out the market’s need to increase productivity without reaching full unemployment (without the guarantee of Universal Basic Income).
Unlike what most sci-fi writers envisioned:
It is not “Skynet becoming self aware” that we need to worry about, but the harm it could do to the economy and to people’s quality of life.
https://deepmind.com/research/publications/2019/playing-atari-deep-reinforcement-learning
https://theconversation.com/stephen-hawking-warned-about-the-perils-of-artificial-intelligence-yet-ai-gave-him-a-voice-93416
https://www.washingtonpost.com/business/whats-a-pingdemic-and-why-is-the-uk-having-one/2021/08/02/df6e3c24-f3ac-11eb-a636-18cac59a98dc_story.html
https://www.bloomberg.com/news/articles/2021-09-28/softbank-backed-autostore-to-raise-315-million-in-oslo-ipo
https://www.aei.org/economics/what-atms-bank-tellers-rise-robots-and-jobs/