Soapbox

Sussing out artificial intelligence’s impact on your job

BRETT, the Berkeley Robot for the Elimination of Tedious Tasks, which uses a new approach to training robots to match human dexterity and speed, at UC Berkeley, May 20, 2015. The new approach includes a powerful artificial intelligence technique known as deep learning to achieve major advances in both computer vision and speech recognition.
BRETT, the Berkeley Robot for the Elimination of Tedious Tasks, which uses a new approach to training robots to match human dexterity and speed, at UC Berkeley, May 20, 2015. The new approach includes a powerful artificial intelligence technique known as deep learning to achieve major advances in both computer vision and speech recognition. New York Times file

Coming soon to your workplace, if not already there: robots, artificial intelligence, machine learning and related technology to pull things together. How do we prepare for that?

The White House recently issued a report, “Preparing for the Future of Artificial Intelligence.”

In the near term, the report sees the “development of intelligent systems that work well as helpers, assistants, trainers, and teammates of humans.” In the long term, there will be systems that quickly race “far ahead of humans in intelligence.”

The report’s bottom line is that there will be “a new age in the global economy” and “it is to be expected that machines will reach and exceed human performance on more and more tasks.”

These systems are driven and enabled by artificial intelligence, or AI, but you will hear them described as automation, robotics, machine learning, or something cozier like the newly familiar names of Siri, Echo or Home. The White House report notes the lack of a universally accepted definition of AI, but says the core objective of AI research and application is to “automate or replicate intelligent behavior.”

To ensure these intelligent systems are designed to operate safely and ethically, the report calls for the federal government to monitor developments worldwide and assess risks related to AI.

While there are expectations of increased productivity and dramatic wealth creation as AI spreads in the work world, it is also likely there will be negative effects hitting hardest at lower-wage jobs and increasing inequities between less-educated and more-educated workers.

“An AI-enabled world,” the report states, “demands a data-literate citizenry that is able to read, use, interpret and communicate about data, and participate in policy debates about matters affected by AI.”

The report speaks to public policy “ensuring that workers are retrained and able to succeed in occupations that are complementary” and adds that public policy “can also ensure that the economic benefits created by AI are shared broadly.”

That last reference to shared benefits reflects the thinking of some, including Obama, who suggest consideration of a universal basic income as a means of distributing the excessive wealth expected to be created by AI’s invasion of the workplace. Why? It would be a means of compensating the likely unemployment consequences of the AI impact on the workforce.

It is natural to start thinking about the loss of jobs and the elimination of certain occupations. A recent research paper by McKinsey & Company says the workplace impact will be nuanced, but “will affect portions of almost all jobs to a greater or lesser degree, depending on the type of work they entail.” The most accurate way to analyze the impact is to look at work activities rather than occupations.

“Currently demonstrated technologies,” McKinsey found, “could automate 45 percent of the activities people are paid to perform and that about 60 percent of all occupations could see 30 percent or more of their constituent activities automated, again with technologies available today.” Most susceptible to automation are food service and accommodations, manufacturing and retailing. All involve “predictable physical activities.”

Occupations involving the collection and processing of data fall into a middle range of potential for automation. While they fall into a low potential for automation, 27 percent of education activities, mostly those outside the classroom, and about 35 percent of health care activities have a technical potential for automation under current technologies.

Jobs in general are likely to become more complex and require more education but will also have more technology available for support. That transformation is likely to pour salt into the existing socioeconomic wounds reflected in our presidential election. Many are unprepared and many could be left behind in this predictable future of work.

The challenges to our current system of education, which will need a big rethink, and the existing economic safety net will be formidable. An unusual collaboration among business, labor, government and academia will be needed to frame new policy to help us adapt to new realities.

John M. Hein is a public policy consultant and the former director of governmental relations for the California Teachers Association. His most recent piece, “Driverless vehicles propel us into an uncharted future,” ran on Oct. 2. j-hein@sbcglobal.net

  Comments