Monday, June 7, 2021

Artificial Intelligence, the American Worker and a Universal Basic Income

With the growing talk of official implementation of a universal basic income picking up momentum in the post-pandemic world, a recent development gives us a sense of why governments will be under pressure to ensure the financial security of their citizens.


According to Nation's Restaurant News, McDoncald's CEO Chris Kempczinski told an investor conference being held by AllianceBernstein Strategic Decisions that McDonald's is testing automated voice order-taking at ten drive-through restaurants in the Chicago area.  


Let's look back to September 2019, when McDonald's moved into the voice-based and multi-item conversational ordering technology sector though its acquisition of Apprente as shown here:


Here is a quote from McDonald's press release dated September 10, 2019 outlining the companies rationale for acquiring Apprente:


"Building our technology infrastructure and digital capabilities are fundamental to our Velocity Growth Plan and enable us to meet rising expectations from our customers, while making it simpler and even more enjoyable for crew members to serve guests," said Steve Easterbrook, (then) President and Chief Executive Officer, McDonald's Corporation. "Apprente's gifted team, and the technology they have developed, will form McD Tech Labs, a new group integrated in our Global Technology team that will take our culture of innovation one step further.""


The Apprente team will be the founding member of a new, internal group inside of McDonald's Global Technology Team called McD Tech Labs.  This acquisition builds on the April 2019 acquisition of Dynamic Yield, an artificial intelligence company.  According to Dynamic Yield's CEO, the acquisition:


' great news for Dynamic Yield and McDonald’s customers, as well as personalization practitioners everywhere. Dynamic Yield will play a critical role in McDonald’s digital transformation, allowing it to become even more focused on the customer by deploying our technology in outdoor digital Drive-Thru menu displays, as well as other digital customer experience touchpoints, such as self-order kiosks and the McDonald’s Global Mobile App."


Let's go back to the recent revelations made by McDonald's current President and CEO Chris Kempczinski.  He told the investor conference that, while the automated voice-ordering system was 85 percent accurate and can take about 4/5ths of orders in all tests, that it will be some time before the company plans to roll out the system in its 13,673 restaurants (as of December 31, 2020) across the United States and, one has to presume, its 39,198 restaurants around the world. 


Here are some additional quotes from Chris Kempczinski:


"There is a big leap between going from 10 restaurants in Chicago to 14,000 restaurants across the U.S. with an infinite number of promo permutations, menu permutations, dialect permutations, weather — I mean, on and on and on and on.  Do I think in five years from now you're going to see a voice in the drive-thru?  I do, but I don't think that this is going to be something that happens in the next year or so...."


One of the things that we've learned in our 10 restaurants that we've done it is: How do you train a crew to actually not want to jump in as soon as they hear a question or a pause?  We've had to do a little bit of training of ‘just keep your hands off the steering wheel, let the computer do its work.’”


Kempczinski also noted that staff needed to develop a comfort level with the AI technology: 


We weren't getting enough of the orders to be actually able to be processed through the voice recognition technology, because as soon as there was a question or a hiccup, the crew had a tendency to just want to jump in. And it took a little bit of time to actually learn to trust the technology.

Yes, by all means, trust (and learn to love) the technology that is going to put you out of a job.


As shown on this graphic from McDonald's, over the past several years, the company has increased expenditures on technology at the same time as it is reducing General and Administrative (G&A) spending:

Now, let's take a quick look at McDonald's wages for its front line workers (in the United States in U.S. dollars):


Here are the 2020 salaries for the company's Named Executive Officers:


If you were to take the salary of a fast food attendant of $9.83 per hour for a 40 hour week, their before tax income would be $393.20 for a week's work or $20,446.40 for a year's work (no holidays).  At this level, a fast food attendant at McDonalds would have to work for 4.92 years to earn the monthly compensation  (including perquisite and benefits) of $100,560.67 earned by the company's CEO.  And yet the company is looking to replace its front line, low-paid workforce with an AI-equipped automatic ordering system, putting untold thousands of workers out of jobs.  

It's no wonder that there is increasing discussion about a universal basic income.  With technology, particularly artificial intelligence, displacing workers on the low end of the pay scale, the American work force is on the cusp of substantial and permanent changes that will benefit only the few at the top of the salary pyramid.  After all, it's all about profits, isn't it?

1 comment:

  1. David Goldman at Asia Times has an interesting essay on 5G and AI. He claims that China has 70% of all installed 5G base stations (ca. 800,000) and will double the number of base stations in the next few years. It is all Huawei equipment so it is 5 times faster than the enhanced 4G being installed by AT&T, Verizon, et al., with much wider bandwidth and much lower latency.

    China is also heavily investing in AI to run on these networks, and it is installing the two in its automated factories. The basic idea is that with very high speed, low latency networks, robots with AI can do self-learning and self-coordination, taking people out of the loop.

    China appears to be a few laps ahead of the US in 5G and AI. They may be on the verge of a permanent competitive advantage in manufacturing.

    Military applications are close.