Man vs Machine: a gloomy outlook?

The economic recovery has hit a wall. During the February 17th, 2010 anniversary of the signing of the stimulus bill, President Obama addressed the nation prompting many to believe the economic stimulus plan provided the resuscitation needed to help restore jobs in America. "So far the Recovery Act is responsible for the employment of about 2 million Americans who would otherwise be out of work, said Obama. The Recovery Act is on track to save or create another 1.5 million jobs in 2010."

While there’s no doubt that the U.S. economy is recovering and should continue growing through next year, the unemployment rate is expected to remain higher than usual and will take some time before the jobs lost in 2009 are replaced.

Just the other day Hewlett-Packard announced plans to lay off about 9,000 people or 3% of its work force. The workers are being replaced by super intelligent software that will reduce operating expenses, increase productivity and save HP millions of dollars. That’s great for Hewlett-Packard, but it sounds like more out of work unhappy people to me.

The media regularly reports how the economy continues to struggle as a result of high unemployment, and here goes Hewlett Packard using job automation technology to do the work humans could do. But then again, who can blame HP? All businesses want greater productivity while using a smaller work force to gain a competitive advantage. What better way to achieve that goal than through the use of super computers and intelligent software?

Where are we headed?

Job automation can be defined as the replacement of human workers by machines for certain tasks. The phenomenon began during the Industrial Revolution and has been transforming society ever since. You may think using machines in place of humans is a great way to boost productivity, increase a company’s bottom-line as well as relieving people from unsafe or tedious tasks. But the catch is of course the more efficient the machine the more human jobs lost. If automation prevents the economy from creating new jobs then the worker not hired is as great a burden to the economy as the worker who’s been laid off or fired.

It is evidence that job automation technology will soon advance to the point where certain repetitive jobs will become scarce to humans. Before the introduction of those unmanned predator drones the military uses in Iraq, I tried to convince my father that at some point robotics could advance to where automated technology would replace the need for both low and high skilled workers. While my father insisted even the most intelligent machines would be unable to think in advance and anticipate a person’s needs, I believed otherwise.

Fast forward some 25 years to Virtuoz, a San Francisco based company that created Rapid Adapt. Rapid Adapt is a type of software that allows interactive voice response systems (press one for yes, press two for no) to learn and become smarter based on every customer’s experience. It’s an advanced version of software similar to what IBM developed a few years ago for a call center managed by PNC Global. You can bet that Cray Systems, Google and others that we don’t know about yet are busy at work creating complex machines that can learn and replicate human thought. We’re probably just an algorithm or two away from producing Joshua, the fictional supercomputer from the movie War Games that taught itself the concept of futility.

Eliminating inefficiencies

I’ll admit a computer that can “reason and think” sounds exciting, but the downside to the digital age is that today's new industries have comparatively fewer jobs for the unskilled or semiskilled worker, and our unwillingness to acknowledge this real issue may derail any hope for a successful job market recovery.

Maybe some of you reading this are familiar with Moore’s Law. In 1965 Intel co-founder Gordon Moore predicted that the number of transistors on a computer chip would double yearly. So while we sit patiently awaiting the stimulus packages to produce something close to full employment, technology is moving full steam ahead. Some computer designers have already abandoned silicon chips, preferring instead to use and experiment with DNA chips which are capable of storing extraordinary amounts of data and perform calculations a billion times faster than what is on the market today. This shift in integrated circuitry is a giant step forward in computer science, and in man’s attempt to develop artificial intelligence.

Automation will become the trend

In a book written by Martin Ford titled The Lights in the Tunnel: Automation, Accelerating Technology and the Economy of the Future, Mr. Ford confronts the implications of J.A.T (job automation technology) and argues that it will continue advancing aggressively into the future resulting in significant and increasing structural unemployment. I believe as Mr. Ford that robots and computers will eventually replace humans in many industries over the next 10 to 20 years. I want to believe that for every worker taken off the assembly line another will be added to the “robotic maintenance crew” while another is added to the “automated consulting team” thereby making it impossible to do away with human jobs entirely.

However, while there will obviously always be a need for people to work in some capacity, it’s a fact that new employment sectors tend to be expensive and have fewer people qualified to lead the way in the beginning stages. This leaves the door open to the disruptive force of technology to once again reduce manual intervention by human workers, as CEO’s favor intelligent software in place of the engineering graduate with no experience. While it might take the graduate student months to learn a set of complex skills, a robot can simply be programmed in seconds to do the same job at half the expense.

To see the difference in employment between a regular industry and a new technology industry, compare Wal-Mart (more that 2 million “associates” worldwide and revenue of about $180,000 per worker) with the almighty Google (20,000 full-time employees with over a million dollars per worker). Is there any question that Wal-Mart, a company known for offering low prices; won’t resemble Google in the near future? The corporate trend going forward will be to harvest the vast potential that technology has to offer while using fewer employees to do so.

A competitive edge

The once mighty Blockbuster Video is now paying the ultimate late fee because it failed to keep up with the advancement of technology. Blockbuster employed thousands of people in almost 10,000 retail locations at one time. Then Netflix entered the picture with a few highly automated distribution centers and the game changed overnight. Streaming video on-line has become the new trend as opposed to brick and mortar stores because it’s inexpensive, quick and convenient and unfortunately for Blockbuster that reality has pushed the company toward near bankruptcy. Netflix represents a perfect example of a corporation that utilized the vast potential technology has to offer while using fewer employees to do so.

The genie is out of the bottle

The fact is job automation is here to stay and will increasingly affect people with MBAs and PhDs (...even in the most unlikely ways). In other words no longer will blue collar, uneducated workers be the sole victims of advanced computer science. A job that at one time required significant training and education will increasingly be done by efficient, cost effective machines. The Stanford MBA grad that once performed statistical analysis for General Electric could be replaced with decision making software capable of using sophisticated algorithms to bolster a company's bottom line.

In the beginning when it was called the “World Wide Web”, chat rooms relied on flesh-and-bone moderators as public deterrents for posting inappropriate material on-line. Nowadays companies such as Facebook, while having a review team on staff, utilize a number of sophisticated algorithms that detect inappropriate activity among users by tracking behavior and flagging activity deemed suspicious. The site analyzes member actions and compares them to others.  Users who experience a significant number of declined friend requests for example are flagged, as are users who persistently and aggressively pursue relationships with children and young women. In some instances people sending out too many messages at once will receive a warning with a captcha to determine if they are real or simply a computer sending spam. Captcha stands for Completely Automated Turing Test To Tell Computers and Humans Apart (It’s the distorted text where you have to enter a set of letters before you can proceed on a web page).

No turning back

With computer science accelerating at a constant rate, middle class Americans with and without higher education will discover fewer opportunities in the job market as technology enables computers and machines to perform sophisticated analysis as well as repetitive, routine task. Now I’m sure my assumptions will not go unchallenged. I acknowledge some jobs are here to stay and may never be automated, or even if they could be, people will always want humans in those positions. Doctors, painters, grief counselors, elected officials; they all provide a human touch that probably can’t be replicated by even the most sophisticated machine. If such a “Job Terminator” ever comes to exist, then I would imagine the economy may be the least of our concerns.