Why do so many politicians and ideologues dislike Google,
Facebook, and Amazon? Are they too monopolistic, and are they using our data
ethically? Also, how can we make broadband more accessible for rural America, and
what policies should we put in place in order to fully benefit from the rise of
artificial intelligence? Will Rinehart and I recently discussed these questions.

Will is the director of Technology and Innovation Policy at
the American Action Forum, where he specializes in telecommunication, Internet,
and data policy. He is also a Frédéric Bastiat Fellow at the Mercatus Center,
and was previously a research fellow at Tech Freedom.

Below is an abbreviated transcript of our conversation. You can read our full discussion here. You can also subscribe to my podcast on iTunes or Stitcher, or download the podcast on Ricochet.

Pethokoukis: Big Tech companies like Apple, Amazon, Google, and Facebook seem really unpopular among political activists on both sides, but consumer surveys show that these companies are still really popular among most Americans. Where does this discrepancy come from?

Rinehart: There are a couple reasons. There’s the
relationship of Facebook with the 2016 election, and this question of fake
news, or whether or not democracy is being harmed by the fact that we have
Twitter, Google, and Facebook.

There’s also the competition concern. These are large
companies, and they don’t seem to have natural competitors for their primary
functions. There doesn’t seem to be a major competitor to Facebook, even though
Facebook is somewhat competitive in many other spaces. And, Google doesn’t
really have a natural competitor in search, but it has competition with the
Android system, and it’s competitive against AWS and their cloud services.

So this “techlash” is partly political, and partly a result
of changing nature of competition.

Google has about 90
percent of search, Amazon dominates online commerce, and Facebook owns so many
of the big social media companies. Are these companies dominating their markets
and shutting out competition, and is this a problem?

What do you mean by competition? As an economist, you worry
about concentration as a kind of “first test” as to whether you should be
concerned about an industry. When it comes to concentration, again, the numbers
don’t necessarily evince the fact that these companies are anti-competitive.

Google, for example, makes around five hundred changes to
their algorithm every day — they’re still chasing high quality search results
regardless of the fact that they are obviously a dominant player in search. To
me, that’s a sign of competition — of a company that feels like it is under
competitive threat.

When we normally talk about questions of competition, it
really has universally been related to price. So we’re looking for the reason
why you’d have fewer competitors: You can increase price. And when it comes to
Google’s search, that doesn’t really exist.

Another big concern
is that these companies use your data as part of a targeted-advertising
business model. What are your thoughts on that? Should, for instance, Google or
Facebook be paying me for using my data? It’s obviously very valuable to them,
so why am I not getting a check?

You’re already getting an implicit benefit. There’s really interesting work that Erik Brynjolfsson has done on this. When he asks an individual, “How much would you be willing to be paid to not use Google’s search anymore?” it’s typically around $1,800 a year. I mean, that’s a potentially big check, or at least an implicit value you’re placing on Google’s search.

When you typically ask people, “Do you want to be paid for
your data?” and then you start asking follow-on questions — “Well, would you be
willing to pay Facebook for the use of its services?” they say, “Yes, I want to be paid for my data being
collected, but also I don’t want to have to pay for the service.” So there’s a
lot of conflict here, and it’s understandable — everyone wants free money.

How big of a problem
is broadband access in rural America, and what should we be doing about it?

Typically, rural areas have downtown areas with good internet
access. But when you get just a mile outside of them, they sometimes don’t.

States clearly are trying to solve this problem. You’ve seen
in the last five years that localities are trying to do targeted grants and
loans to get broadband into the regions that really need it. And the federal
government is doing a lot: The FCC still spends $8 or $9 billion per year on
various broadband development programs. There’s a lot that’s being done.

What you’re going to start seeing, especially in some of
these rural regions, is localities working with some of these ISPs to figure
out where broadband needs to be located, and how to help with it. There’s a
whole range of options that localities are trying to pursue, because, there’s a
whole bunch of different ways to do this.

I’m really worried
that China’s going to win the AI race. Are you worried about that, and what
should the federal government be doing to help advance that technology, if
anything?

As with every previous technology, the AI race hasn’t
necessarily been about the race to the quickest technology development itself.
It has always been about applications, new business methods, and the
application of education to business.

When you talk to people that are concerned about the AI question, the biggest concern that almost everyone cites is education. The algorithms are there, the hardware is accessible and cheap, and those are the two things that have occurred in the last few years that make AI possible. But the thing that really makes it possible that we need to get right is education. Immigration is also important — some of the best engineers that work in the United States have come from other countries. They want to settle here because we have good jobs, and there are often very nice places to work at if you’re working on AI applications. That is what we really need to be focused on — education and immigration.

The post 5 questions for Will Rinehart on Big Tech, rural broadband, and the AI race appeared first on American Enterprise Institute – AEI.