You say that Baidu already has more than a dozen mobile
apps used by 300 million people outside China every month.
What is your strategy for other markets?
We are very focused on mobile. Our apps are popular in India, Brazil,
parts of Southeast Asia like Indonesia. Those are places where the
market is very similar to China, but maybe [China as it was] a few
years ago—large populations that are mobile-centric. And that gives
an advantage over U.S. companies.
Will you offer products in the U.S.?
While we have over 20 million users in the U.S., our focus is on R&D and
talent. Right now PhDs from the best Chinese universities are on par with
the top U.S. graduates. We have a lab here in the U.S. because we also
need more senior talent. The AI team in the U.S. has already developed
a new speech-recognition system that is playing a role in our product.
You have spent a lot on artificial-intelligence research. How
does that translate into new revenue?
AI will be like electricity, in that it will drive a lot of new business to many industries. We have built a platform we and partners can build on to offer services in anything—health care,
education, logistics. As one example, we have a new partnership
with Bosch around using our high-definition maps for autonomous
and semi-autonomous cars. Many of these [services] will probably be bigger than search, which will continue to be our core
business for many years and is also being transformed by AI.
Your chief scientist Andrew Ng, who previously helped
establish Google’s machine-intelligence research, left the
company recently. Is this a long-term setback?
Andrew is a great guy, and it’s a loss to Baidu here in the U.S. and
to the wider company. But he built a large AI team with a lot of talent, and Baidu has a very deep bench of leaders. Even in the U.S.
labs, more than 60 percent of employees—on autonomous driving, security, our advertising platform—worked for somebody else.
Why would top talent in artificial intelligence want to work
for Baidu’s U.S. lab over that of an American company?
We allow researchers and engineers to make a bigger impact with
their work. You get to be part of the two most exciting and dynamic
technology ecosystems on Earth: Silicon Valley and the Internet
economy of China. The Chinese market gives researchers access
to a massive amount of data and a consumer base incredibly quick
to adopt new technology.
Baidu is developing and testing autonomous cars in both
the U.S. and China. The leading projects in this area are
almost all based in the U.S. Where will this technology be
In China the roads and traffic are more complex than in the U.S.
But the first batch of applications will be in restricted areas, and it
could be easier in China because regulators are more flexible and
open-minded. Mayors of many different places want us to try on
their roads and are willing to build out special areas for us. A couple of major ports in China want to use our technology on trucks.
Are there areas where you are ahead of U. S. competitors?
I think we deploy new algorithms into our products much faster. The
Chinese market is hypercompetitive, and to avoid being beaten by
competitors, the cycle of [translating new technology into prod-ucts] is much shorter than in places like the U.S. Startups in China
are even faster than us at making new AI algorithms into products,
and that’s something to look out for.
How are Chinese tech companies doing overseas so far?
We’re all just beginning. None of the biggest Chinese Internet companies has proven success outside of China, but I think in the next
five years, 10 years, you will see those companies become global.
Alibaba, Tencent, Baidu, and others—we are all deploying teams
and building up talent. Next will come more products and many
Google and Facebook are household names around the
world. Baidu? Not yet. Ya-Qin Zhang, president of China’s
leading search business, says Chinese companies can
become worldwide Internet powerhouses too. One of the
biggest fish in China’s market of 730 million Internet users,
Baidu is trying to open new revenue streams both domestically
and abroad by investing heavily in artificial intelligence.
The company employs more than 1,700 AI researchers,
including some at a Silicon Valley research center opened
in 2014, and was chosen by the Chinese government to
run a new national lab intended to make the nation more
competitive in machine learning. Zhang even predicts that the
self-driving cars Baidu is developing might be in widespread
public use before those introduced by U.S. competitors.
Domestically, revenue growth in its existing search and
ads business wilted in 2016 as smartphone sales slowed
in China and rivals including Tencent, owner of the popular
mobile messaging and e-commerce platform WeChat,
Baidu’s reputation and ad business were damaged as
well by a scandal in which a student died from cancer after a
sponsored result displayed by the company’s search engine
led him to expensive and questionable treatment.
Zhang, who previously led Microsoft’s research and
business operations in China, took time out from meeting
staff at Baidu’s U.S. R&D center in Silicon Valley to meet with
MIT Technology Review’s San Francisco bureau chief, Tom
Simonite. An edited version of their interview is below.