The AI revolution may not be simply a jobs destroyer

For years, conventional wisdom has suggested that artificial intelligence (AI) is poised to tear through traditional industries like a virus, displacing established human workers with smart software in the process. With such a reputation for killing jobs, it’s encouraging that a new report from career site Paysa paints a very different picture for AI: Significant job growth with dramatic investments in hiring for at least the next five years.

According to Paysa, technology companies are doubling down on AI research, with more than $650 million being added to corporate headcount for additional AI talent this year alone. Twenty top employers are spending an average of $33 million to attract new talent for this developing sector.

And that’s just an average: Amazon (AMZN) leads the way with a budget of about $228 million for new engineers and scientists specializing in AI. Google (GOOG) comes in second at $130 million.

Interestingly, IBM (IBM) trails the pack at just $5 million, suggesting perhaps that the company isn’t pinning its future on AI systems in the same way as internet companies like Amazon, Microsoft (MSFT), Google and Facebook (FB). “What leaps out at me is who’s on the list and who’s not,” said Paysa CEO Chris Bolte. “For example, it’s clear -- and Jeff Bezos confirmed this -- that AI is underpinning everything Amazon is going to do in the future.”

Based on the Paysa report, employers are looking to fill as many as 10,000 positions, making AI look just a little bit like a modern high-tech gold rush. At stake: Science and engineering positions with specialties in deep learning, machine learning, artificial intelligence, computer vision, neural networks and reinforcement learning skills. In the tech sector, said Bolte, “AI is the new talent currency.”

But that suggests a concern: Are 10,000 skilled scientists and engineers ready to step into these roles? 

Probably not, thinks Tim Estes, founder of Digital Reasoning, developer of enterprise-level AI products. And while there will be initial shortages for these kinds of roles, Estes likens AI in 2017 to the “HTML phase” of the internet. “We’ve always solved labor skillset problems with better and simpler tools, and I expect that’s what will happen with AI as well.”

Startups and Silicon Valley aren’t the only ones getting into the AI game. As the top 20 companies in the Paysa report show, more traditional tech companies are trying to acquire AI tech talent as well, and that’s leading to new compensation packages. “We’re seeing that they’re having to change the way they think about employees, changing their comp structure to more like startups to attract that kind of talent,” said Bolte.

And the jobs are by no means evenly distributed across the U.S. Because so much of the investment is happening in a handful of tech companies, it paints a lopsided geographic picture. Far and away, the largest investment is happening in the West, home to companies like Amazon, Google, Microsoft, NVIDIA (NVDA) and Facebook (the top five companies on an AI hiring spree).

About $1.1 billion is currently on the table on the West Coast, thanks almost entirely to San Francisco and Seattle. And it’s no contest: The Northeast comes in a very distant second with $289 million in hiring dollars. The Midwest barely registers, with $12 million.

While AI is indeed adding jobs to the overall workforce -- especially in the short term -- the fact remains that these are highly skilled positions. A survey of open job descriptions demonstrates that 35 percent require a Ph.D., and another 26 percent require a Master’s degree. Less than 20 percent don’t require any kind of college degree. 

That means the tech sector may see a lot of growth among highly degreed engineers as AI systems evolve and find their way into a panoply of industries, but does that suggest the prognosis for workers in those target fields might not be as rosy?

Some experts still predict that AI will take a toll on traditional jobs. “Certain kinds of jobs, especially in retail, just aren’t going to make any economic sense,” predicts Estes. “It’s especially ironic that the living wage movement -- even if it’s morally defensible -- is only going to exacerbate the problem and make traditional jobs an increasingly unsustainable model.”

AI will surely change these jobs, sometimes in unpredictable ways. “In a few years, they will both look very different,” said Estes, “and there will be fewer of them.”

But increasingly, many in the field don’t think AI will be a net job killer at all. “There will be blood -- there will be automation,” said Malcolm Frank, author of “What to Do When Machines Do Everything.” “Most every white collar job will be touched by AI in the next decade.”

But Frank asserted that the unemployment level in 2025 will be roughly what it is today. “People get stuck thinking about jobs holistically, “ he said. “But any job, like nurse, teacher, or janitor, is a collection of tasks. AI will automate some, but not all, of those tasks, allowing humans to do a better job at the things humans do well.”