16 Leaders The EconomistOctober 12th 2019
12 Jim Hackett, the boss of Ford, acknowledges that the industry
“overestimated the arrival of autonomous vehicles”. Chris Urm-
son, a linchpin in Alphabet’s self-driving efforts (he left in 2016),
used to hope his young son would never need a driving licence.
Mr Urmson now talks of self-driving cars appearing gradually
over the next 30 to 50 years. Firms are increasingly switching to a
more incremental approach, building on technologies such as
lane-keeping or automatic parking. A string of fatalities involv-
ing self-driving cars have scotched the idea that a zero-crash
world is anywhere close. Markets are starting to catch on. In Sep-
tember Morgan Stanley, a bank, cut its valuation of Waymo by
40%, to $105bn, citing delays in its technology.
The future, in other words, is stuck in traffic. Partly that re-
flects the tech industry’s predilection for grandiose promises.
But self-driving cars were also meant to be a flagship for the pow-
er of ai. Their struggles offer valuable lessons in the limits of the
world’s trendiest technology.Hit the brakes
One is that, for all the advances in machine learning, machines
are still not very good at learning. Most humans need a few dozen
hours to master driving. Waymo’s cars have had over 10m miles
of practice, and still fall short. And once humans have learned to
drive, even on the easy streets of Phoenix, they can, with a little
effort, apply that knowledge anywhere, rapidly learning to adapt
their skills to rush-hour Bangkok or a gravel-track in rural
Greece. Computers are less flexible. airesearchers have expend-ed much brow-sweat searching for techniques to help them
match the quick-fire learning displayed by humans. So far, they
have not succeeded.
Another lesson is that machine-learning systems are brittle.
Learning solely from existing data means they struggle with situ-
ations that they have never seen before. Humans can use general
knowledge and on-the-fly reasoning to react to things that are
new to them—a light aircraft landing on a busy road, for in-
stance, as happened in Washington state in August (thanks to
humans’ cognitive flexibility, no one was hurt). Autonomous-
car researchers call these unusual situations “edge cases”. Driv-
ing is full of them, though most are less dramatic. Mishandled
edge cases seem to have been a factor in at least some of the
deaths caused by autonomous cars to date. The problem is so
hard that some firms, particularly in China, think it may be easi-
er to re-engineer entire cities to support limited self-driving
than to build fully autonomous cars (see Business section).
The most general point is that, like most technologies, what is
currently called “ai” is both powerful and limited. Recent pro-
gress in machine learning has been transformative. At the same
time, the eventual goal—the creation in a machine of a fluid,
general, human-like intelligence—remains distant. People need
to separate the justified excitement from the opportunistic
hyperbole. Few doubt that a completely autonomous car is pos-
sible in principle. But the consensus is, increasingly, that it is not
imminent. Anyone counting on ai for business or pleasure
could do worse than remember that cautionary tale. 7I
magine a central bank tweeting that, yes, there are rumours
of “certain” banks facing deposit runs but “there is no need to
panic”. Would you feel reassured? That is the unenviable posi-
tion Indians found themselves in last week as a financial storm
rumbled on in the world’s fifth-biggest economy with no sign of
the authorities getting a firm grip. In the latest fiasco a co-oper-
ative bank, pmc, is accused of fraud, prompting depositors to
yank their cash out. Meanwhile shares in Yes Bank, a private
lender, have collapsed by 40% in the past month
as rumours swirl. These are not isolated inci-
dents. Roughly a third of the financial system is
on crutches or under suspicion. Dazed by the
scale of the task, the government and the Re-
serve Bank of India (rbi) are dithering. Until
they act, India’s economy will not perk up—and
there is a danger of a full-blown crisis.
The origins of this mess go back to 2005. In
the first phase conventional banks, which control about four-
fifths of the system’s assets and are mostly state-run, lent too
freely to infrastructure and industrial projects, sometimes ones
backed by well-connected tycoons. The plight today is a continu-
ation of the second phase: a boom-and-bust in lightly regulated
shadow banks, which control the remaining fifth of the system.
The danger grew in 2016 when the government temporarily abol-
ished large banknotes, leading many people to deposit money in
banks and mutual funds. These, in turn, used the windfall tomake loans to shadow banks, which went on their own lending
binge, often using the money to finance property projects.
Today the financial system is stuffed with bad debts. Perhaps
a tenth of loans are dud, maybe more. The shadow banks are vul-
nerable because they use short-term debt (rather than ordinary
deposits, which they are mostly restricted from raising) to fund
long-term loans of their own. There is also an undercurrent of
fraud and bogus accounting. In 2018 Punjab National Bank said
that a diamond dealer had stolen $2bn from it.
Later that year il&fs, a big shadow bank with
government links, collapsed. Credit-rating
agencies have been giving high ratings to flaky
firms. With suspicion rife, a handful of shadow
banks face a severe funding squeeze, and the en-
tire financial sector is wary of lending. As a re-
sult credit is growing at near its slowest pace in
20 years. The ripple effect has stalled building
projects, starved wholesalers of loans to buy inventory and pre-
vented farmers from borrowing to buy tractors and motorbikes.
The response of Narendra Modi’s government and the rbihas
so far been halting. The government has repeatedly but belatedly
pumped inadequate sums of capital into the state banks, and
promised to merge some of them. On September 20th it slashed
corporate taxes to try to revive animal spirits. The rbi, mean-
while, has cut interest rates five times this year. Presumably they
hope that this will be enough to boost the economy, while the bigA big stink on the brink
India’s future should be bright. A rotten financial system could ruin itIndia’s economyIndia’s Yes Bank
Share price, rupee2018 20190100200300400