Artificial
Intelligence Swarms Silicon Valley on Wings and Wheels
By
JOHN MARKOFF JULY 17, 2016 SUNNYVALE, Calif. —
For
more than a decade, Silicon Valley’s technology investors and entrepreneurs
obsessed over social media and mobile apps that helped people do things like
find new friends, fetch a ride home or crowdsource a review of a product or a
movie. Now Silicon Valley has found its next shiny new thing. And it does not
have a “Like” button. The new era in Silicon Valley centers on artificial
intelligence and robots, a transformation that many believe will have a payoff
on the scale of the personal computing industry or the commercial internet, two
previous generations that spread computing globally.
Computers
have begun to speak, listen and see, as well as sprout legs, wings and wheels
to move unfettered in the world. The shift was evident in a Lowe’s home
improvement store here this month, when a prototype inventory checker developed
by Bossa Nova Robotics silently glided through the aisles using computer vision
to automatically perform a task that humans have done manually for centuries.
The robot, which was skilled enough to autonomously move out of the way of
shoppers and avoid unexpected obstacles in the aisles, alerted people to its
presence with soft birdsong chirps. Gliding down the middle of an aisle at a
leisurely pace, it can recognize bar codes on shelves, and it uses a laser to
detect which items are out of stock.
Silicon
Valley’s financiers and entrepreneurs are digging into artificial intelligence
with remarkable exuberance. The region now has at least 19 companies designing
self-driving cars and trucks, up from a handful five years ago. There are also
more than a half-dozen types of mobile robots, including robotic bellhops and
aerial drones, being commercialized. “We saw a slow trickle in investments in
robotics, and suddenly, boom — there seem to be a dozen companies securing
large investment rounds focusing on specific robotic niches,” said Martin
Hitch, chief executive of Bossa Nova, which has a base in San Francisco.
Funding
in A.I. startups has increased more than fourfold to $681 million in 2015,
from $145 million in 2011, according to the market research firm CB Insights.
The firm estimates that new investments will reach $1.2 billion this year, up
76 percent from last year. “Whenever there is a new idea, the valley swarms
it,” said JenHsun Huang, chief executive of Nvidia, a chip maker that was
founded to make graphic processors for the video game business but that has
turned decisively toward artificial intelligence applications in the last year.
“But you have to wait for a good idea, and good ideas don’t happen every day.”
By
contrast, funding for social media startups peaked in 2011 before plunging.
That year, venture capital firms made 66 social media deals and pumped in $2.4
billion. So far this year, there have been just 10 social media investments,
totaling $6.9 million, according to CB Insights. Last month, the professional
social networking site LinkedIn was sold to Microsoft for $26.2 billion,
underscoring that social media has become a mature market sector.
Even
Silicon Valley’s biggest social media companies are now getting into artificial
intelligence, as are other tech behemoths. Facebook is using A.I. to improve
its products. Google will soon compete with Amazon’s Echo and Apple’s Siri,
which are based on A.I., with a device that listens in the home, answers
questions and places ecommerce orders. Satya Nadella, Microsoft’s chief
executive, recently appeared at the Aspen Ideas Conference and called for a
partnership between humans and artificial intelligence systems in which
machines are designed to augment humans.
The
auto industry has also set up camp in the valley to learn how to make cars that
can do the driving for you. Both technology and car companies are making claims
that increasingly powerful sensors and A.I. software will enable cars to drive
themselves with the push of a button as soon as the end of this decade —
despite recent Tesla crashes that have raised the question of how quickly human
drivers will be completely replaced by the technology.
Silicon
Valley’s new A.I. era underscores the region’s ability to opportunistically
reinvent itself and quickly follow the latest tech trend. “This is at the heart
of the region’s culture that goes all the way back to the Gold Rush,” said Paul
Saffo, a longtime technology forecaster and a faculty member at Singularity
University. “The valley is built on the idea that there is always a way to
start over and find a new beginning.”
The
change spurred a rush for talent in A.I. that has become intense. “It’s
ridiculous,” said Richard Socher, chief scientist at the software maker
Salesforce, who teaches a course at Stanford on a machine intelligence
technique known as deep learning. “The number of people trying to get the
students to drop out of the class halfway through because now they know a
little bit of this stuff is crazy.”
The
valley’s tendency toward reinvention dates back to the region’s initial
emergence from the ashes of a deep aerospace industry recession as a consumer-electronics
manufacturing center producing memory chips, video games and digital watches in
the mid1970s. A malaise in the personal computing market in the early 1990s
was followed by the World Wide Web and the global expansion of the consumer
internet. A decade later, in 2007, just as innovation in mobile phones seemed
to be on the verge of moving away from Silicon Valley to Europe and Asia, Apple
introduced the first iPhone, resetting the mobile communications marketplace
and ensuring that the valley would — for at least another generation — remain
the world’s innovation center.
In
the most recent shift, the A.I. idea emerged first in Canada in the work of
cognitive scientists and computer scientists like Geoffrey Hinton, Yoshua
Bengio and Yann LeCun during the previous decade. The three helped pioneer a
new approach to deep learning, a machine learning method that is highly
effective for pattern recognition challenges like vision and speech. Modeled on
a general understanding of how the human brain works, it has helped
technologists make rapid progress in a wide range of A.I. fields.
How far
the A.I. boom will go is hotly debated. For some technologists, today’s
technical advances are laying the groundwork for truly brilliant machines that
will soon have human-level intelligence. Yet Silicon Valley has faced false
starts with A.I. before. During the 1980s, an earlier generation of
entrepreneurs also believed that artificial intelligence was the wave of the
future, leading to a flurry of startups. Their products offered little
business value at the time, and so the commercial enthusiasm ended in
disappointment, leading to a period now referred to as the “A.I. Winter.”
The
current resurgence will not fall short this time, said several investors, who
believe that the economic potential in terms of new efficiency and new
applications is strong. “There is no chance of a new winter,” said Shivon
Zilis, an investor at Bloomberg Beta who specializes in machine intelligence
startups. John Shoch, a veteran venture capitalist at Alloy Ventures in Palo
Alto, Calif., said deep learning has made a difference to the potential success
of A.I. companies. “You get a new set of tools that let you attack a new set of
problems, which let you push the boundary out,” he said. For others, like Jerry
Kaplan, who helped found two A.I. companies in the 1980s — Symantec, which
became a security company, and Teknowledge, which ultimately shut down — the
Valley’s new enthusiasm is troubling because it suggests an unfounded optimism
similar to earlier eras in which the field overpromised and under-delivered.
“Sometimes when I hang around with A.I. enthusiasts here in the valley, I feel
like an atheist at a convention of evangelicals,” he said.
