AI Weekly: Contrary to current fears, AI will create jobs and grow GDP

The inevitable march toward automation continues, analysts from the McKinsey Global Institute and from Tata Communications wrote in separate reports this week.
Artificial intelligence’s growth comes as no surprise — a survey from Narrative Science and the National Business Research Institute conducted earlier this year found that 61 percent of businesses implemented AI in 2017, up from 38 percent in 2016 — but this week’s findings lay out in detail the likely socioeconomic impacts in the coming decade.
The McKinsey models predict that 70 percent of companies will adopt at least one form of AI — whether computer vision, natural language, virtual assistants, robotic process automation, or advanced machine learning — by 2020. And Tata found unbridled enthusiasm among business leaders for an AI-dominated future; in a survey of 120 of them, 90 percent said they expect AI to enhance decision-making.
McKinsey and Tata both contend that’s a good thing.
McKinsey forecasts that AI could contribute an additional 1.2 percent to gross domestic product growth (GDP) for the next 10 years, and that it could furthermore help to capture an additional 20-25 percent in net economic benefits (equating to $13 trillion globally) in the next 12 years.
Tata’s report, meanwhile, says that those gains won’t come at the expense of jobs, but rather will “create new ways of working,” and “new jobs” in companies. (That jibes with a report from Gartner in December 2017, which predicts that AI will create 2.3 million jobs in 2020.)
“Robots and AI are not going to take away this creative, insightful, empathetic aspect of almost every job,” Ken Goldberg, a leading AI researcher and UC Berkeley professor who coauthored the Tata study, wrote.
That’s not to suggest it’ll be smooth sailing. McKinsey lays out the considerable challenges yet to be overcome, which include the human capital required to label training data; the dearth of labeled data; a lack of transparency in AI systems; difficulties in generalizing machine learning models; and the risk of bias.
That’s to say nothing of AI’s public perception problem.
“Policymakers will need to show bold leadership to overcome understandable discomfort among citizens about the perceived threat to their jobs as automation takes hold,” the authors of the McKinsey report wrote, “[and] companies will [need to] be important actors in searching for solutions on the mammoth task of skilling and reskilling people to work with AI.”
And not everyone will share equally in the forthcoming wealth and prosperity.
McKinsey said the portion of jobs that call for “low digital skills” may fall to 30 percent in 2030 from the current 40 percent, as jobs that require higher skills increase to 50 percent from 40 percent. And “leading countries” like the U.S. and China, it notes — which accounted for 66 percent and 17 percent of investments in AI in 2016, respectively — will benefit more than others.
But that hasn’t dampened the C-Suite’s excitement for AI. Eighty percent of leaders surveyed by Tata said that AI could facilitate team “composition,” “organization,” and “communication,” and 93 percent believe it will enhance employee engagement by helping managers better assess skills and suggest activities that can spark creative thinking.
“The prevalent narrative around AI has focussed on a ‘Singularity’ — a hypothetical time when artificial intelligence will surpass humans,” Goldberg wrote for Tata. “But there is a growing interest in ‘Multiplicity’, where AI helps groups of machines and humans collaborate to innovate and solve problems. This survey … reveals that Multiplicity, the positive and inclusive vision of AI, is gaining traction.”
AI applications like IBM’s race-tracking system threaten to upset that “positive and inclusive” vision, of course. But if McKinsey and Tata are right in their predictions, there’s plenty of reason to welcome AI with open arms.
For AI coverage, send news tips to Kyle Wiggers and Khari Johnson — and be sure to bookmark our AI Channel.
Thanks for reading,
Kyle Wiggers
AI Staff Writer
P.S. Please enjoy this video demonstrating Deep Video Portraits, an AI system that modifies a target actor based on the facial and head movement of a source actor.

From VB

GreyOrange raises $140 million as the fulfillment robot industry heats up
Singapore industrial robotics startup GreyOrange has raised $140 million to expedite warehouse fulfillments with machinery.
Read the full story

BMW launches AI assistant that reacts when you’re tired
BMW today announced plans to roll out an intelligent assistant next year that will do the range of things people have come to expect from assistants like play music, give you directions, or respond to random questions. BMW’s new assistant will speak 23 languages and will be available in May 2019.
Read the full story

Microsoft brings Xbox One voice control to Amazon Echo speakers and Windows 10 PCs
Microsoft today introduced Xbox voice control with AI assistants Cortana and Alexa. Xbox One control is available in a range of places where you can find both intelligent assistants, such as Windows 10 PCs, Cortana and Alexa apps on smartphones, Echo smart speakers, and the Harman Kardon Invoke and Sonos One.
Read the full story

IBM collaborated with the NYPD on an AI system that can search for people by race
IBM secretly collaborated with the NYPD to develop a system that could search people by their skin color, according to The Intercept.
Read the full story

Flytrex Golf delivers food and beverages to hungry golfers via drone
Israeli drone company Flytrex is teaming up with King’s Wall Course and Ease Drones to launch what it’s calling the first fully operational “golf course drone delivery system” in the U.S.
Read the full story

IBM’s new system automatically selects the optimal AI algorithm
Martin Wistuba, a data scientist at IBM Research Ireland, described in a recent blog post and accompanying paper an automated AI algorithm selector. He claims it’s 50,000 times faster than other methods with only a small increase in error rate.
Source: VentureBeat