The Risks Digest

The RISKS Digest

Forum on Risks to the Public in Computers and Related Systems

ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator

Volume 31 Issue 19

Saturday 20 April 2019


AA 300 JFK-LAX incident
1983 Soviet nuclear false alarm incident
Dan Jacobson
Contractor identifies new problems with phase 2 of the Silver Line
"Fallible machines, fallible humans"
The Straits Times and Financial Times
A computerized YouTube fact-checking tool goes very wrong: In flaming Notre Dame, it somehow sees 9/11 tragedy
Election systems in 50 states were targeted in 2016
DHS/FBI via Ars Technica
Mysterious operative haunted Kaspersky critics
Samsung's $2,000 folding phone is breaking for some users after two days
Cyberspies Hijacked the Internet Domains of Entire Countries
Man Bites Dog Dept: MSFT supports human rights!!
Microsoft Email Hack Shows the Lurking Danger of Customer Support
As China Hacked, U.S. Businesses Turned A Blind Eye
Wipro customers hacked, says Krebs. Nothing to see here, says Wipro
Facebook has admitted to unintentionally uploading the address books of 1.5 million users without consent
The Guardian
Utah Bans Police From Searching Digital Data Without A Warrant, Closes Fourth Amendment Loophole
AppleWatch or AnkleMonitor: You Decide
Henry Baker
Fintech fiddles as home burns: 97% of apps lack basic security
Info on RISKS (comp.risks)

AA 300 JFK-LAX incident

"Peter G. Neumann" <>
Wed, 17 Apr 2019 15:04:30 PDT
On 10 Apr 2019, an American Airlines Airbus A321 jet `nearly crashed' during
takeoff at JFK.  The wing apparently scraped the ground and hit a sign and
light pole during takeoff, bending the wing.  "We were banking, uncontrolled
bank 45 degrees to the left," a pilot could be heard saying on the air
traffic control audio of the incident.  It was evidently an `uncommanded
roll to the left', with no explanation yet as to the cause.  Although the
plane did manage to take off, it then returned to JFK 28 minutes later.

1983 Soviet nuclear false alarm incident

Dan Jacobson <>
Fri, 12 Apr 2019 11:47:21 +0800
"...the system reported that a missile had been launched from the United
States, followed by up to five more. Petrov judged the reports to be a
false alarm, and his decision to disobey orders, against Soviet military
protocol, is credited with having prevented an erroneous retaliatory
nuclear attack on the United States and its NATO allies that could have
resulted in large-scale nuclear war. Investigation later confirmed that
the Soviet satellite warning system had indeed malfunctioned."

  [In RISKS-3.39, 18 Aug 1986, we had a "Nuclear false alarm" item,
  contributed by Robert Stroud.  That case triggered nuclear attack sirens
  in Edinburgh.  PGN]

Contractor identifies new problems with phase 2 of the Silver Line (WashPost)

Gabe Goldberg <>
Fri, 12 Apr 2019 19:36:28 -0400
The structures that support the Dulles Airport Metro station's glass wall
are cracked and lack proper reinforcement.

Keith Couch, project director for CRC, downplayed the problems at the Dulles
station, saying that officials are working to find a solution. He said the
fact that the problems were discovered before the project was completed is a
sign that the company's quality control program is working. CRC's
inspections and quality control have come under criticism as the project's
problems have mounted.

Project executive director Charles Stark characterized the issues at the
Dulles station as a "workmanship problem."

"QC is working" to detect workmanship problems.

"workmanship" appears in article once, as does "improve"—but referring to
schedule, not workmanship.

The risk? Nothing changing.

"Fallible machines, fallible humans" (The Straits Times and Financial Times)

Richard Stein <>
Wed, 17 Apr 2019 14:04:14 +0800
Robert Wright byline, behind paywalls as:

1) "Fallible machines, fallible humans," via
retrieved on 17APR2019;

2) "Autonomous machines: industry grapples with Boeing lessons" via

The cited news articles discuss technology-dependent systems (medical
infusion pumps, aircraft, industrial robotic manufacturing) and their
dependency on human engagement to monitor activity.

Today's AI cannot independently comprehend context: they can match patterns,
but cannot rationalize the recognized pattern in a way that emulates a
human's mind.

No machine can be programmed today to process contextual awareness and
independently act to preserve and protect human life during an emergency. An
organization or individual expecting this outcome apparently believes that
science fiction is real. They must be disabused of this fallacy.

In the FT and Straits Times articles, Mark Sujan of University of Warwick
asks, "How do we ensure that the system knows enough about the world within
which it's operation? That's a complex thing."

As noted by Don Norman (see for example),
"The real RISK in computer system design is NOT human error. It is designers
who are content to blame human error and thereby wash their hands of

Demonstrating system behavior when subjected to erroneous or negative input
stimulus can reveal more about system safety-readiness and resilience than
demonstration of behavior under nominal stimulus conditions. Anomalous
system states, in a simulator, can instruct and refine operational

Successful and effective system operation depends on informed, trained, and
engaged human oversight. Safety critical system operators must possess
perspicacity. Clear indicators of anomalous behavior, and insightful
operator reaction to them, are essential to ensure a safe outcome.

A computerized YouTube fact-checking tool goes very wrong: In flaming Notre Dame, it somehow sees 9/11 tragedy (WashPost)

Richard Stein <>
Wed, 17 Apr 2019 16:17:13 +0800

"If the algorithm saw a video of tall structures engulfed in smoke and
inferred that it was related to the attack on the World Trade Center, that
speaks well of the state of the art in video system understanding, that it
would see the similarity to 9/11. There was a point where that would have
been impossible.

"But the algorithms lack the comprehension of human context or common sense,
making them woefully unprepared for news events. YouTube, he said, is poorly
equipped to fix such problems now and probably will remain so for years to

"'They have to depend on these algorithms, but they all have sorts of
failure modes. And they can't fly under the radar anymore,' Domingos said.
'It's not just whack-a-mole. It's a losing game.'"

Risk: Brand outrage incidence frequency multiplies with business
accumulation of technical debt.

Election systems in 50 states were targeted in 2016 (DHS/FBI via Ars Technica)

"Peter G. Neumann" <>
Fri, 12 Apr 2019 9:09:05 PDT

*A joint intelligence bulletin (JIB) has been issued by the Department of
Homeland Security and Federal Bureau of Investigation to state and local
authorities regarding Russian hacking activities during the 2016
presidential election.  While the bulletin contains no new technical
information, it is the first official report to confirm that the Russian
reconnaissance and hacking efforts in advance of the election went well
beyond the 21 states confirmed in previous reports.*

Mysterious operative haunted Kaspersky critics (AP)

J Coe <>
Thu, 18 Apr 2019 14:13:57 +0100
The Associated Press has learned that the mysterious man (who said his name
was Lucas Lambert) spent several months last year investigating critics of
Kaspersky Lab, organizing at least four meetings with cybersecurity experts
in London and New York.

Samsung's $2,000 folding phone is breaking for some users after two days (CNBC)

Gabe Goldberg <>
Wed, 17 Apr 2019 19:39:58 -0400
Samsung's Galaxy Fold is already breaking.
Reviewers who got the device are seeing flickering screens. Some think
because a protective film was removed.
But CNBC's unit is also broken and we did not remove the film.

Samsung's $2,000 folding phone is breaking for some users after two days

Gadget gimmick for its own sake? I use two PC monitors for Windows but don't
have windows span their border—bezels would be intrusive. I can't see
using this phone with a single app spanning the displays and am skeptical
about people paying that much for two separate screens—if it even
operates that way. Surprise, the hinge is a likely failure point.

Cyberspies Hijacked the Internet Domains of Entire Countries (WiReD)

Gabe Goldberg <>
Wed, 17 Apr 2019 20:41:13 -0400
The discovery of a new, sophisticated team of hackers spying on dozens of
government targets is never good news. But one team of cyberspies has pulled
off that scale of espionage with a rare and troubling trick, exploiting a
weak link in the Internet's cybersecurity that experts have warned about for
years: DNS hijacking, a technique that meddles with the fundamental address
book of the Internet.

Researchers at Cisco's Talos security division on Wednesday revealed that a
hacker group it's calling Sea Turtle carried out a broad campaign of
espionage via DNS hijacking, hitting 40 different organizations. In the
process, they went so far as to compromise multiple country-code top-level
domains—the suffixes like or .ru that end a foreign web address --
putting all the traffic of every domain in multiple countries at risk.

The hackers' victims include telecoms, Internet service providers, and
domain registrars responsible for implementing the domain name system.  But
the majority of the victims and the ultimate targets, Cisco believes, were a
collection of mostly governmental organizations, including ministries of
foreign affairs, intelligence agencies, military targets, and energy-related
groups, all based in the Middle East and North Africa. By corrupting the
Internet's directory system, hackers were able to silently use "man in the
middle" attacks to intercept all Internet data from email to web traffic
sent to those victim organizations.

Man Bites Dog Dept: MSFT supports human rights!! (Reuters)

Henry Baker <>
Wed, 17 Apr 2019 21:24:08 -0700
  [Once again, I had to carefully check the date on this article to make
  sure that it wasn't April 1st!]

As much as I applaud the zeal of all the newly converted, I'm far too
cynical to believe a word of Brad Smith, given the *second* article about
Microsoft, below.  Perhaps St. Augustine's prayer is more appropriate for
Microsoft: "Please God, make me good, but not just yet".

My prayer for Microsoft: "May the Farce be with you!" *

(* See below.)

Microsoft turned down facial-recognition sales on human rights concerns

Joseph Menn   April 16, 2019 / 11:33 PM / Updated a day ago

PALO ALTO (Reuters) - Microsoft Corp recently rejected a California law
enforcement agency's request to install facial recognition technology in
officers' cars and body cameras due to human rights concerns, company
President Brad Smith said on Tuesday.

Microsoft concluded it would lead to innocent women and minorities being
disproportionately held for questioning because the artificial intelligence
has been trained on mostly white and male pictures.

AI has more cases of mistaken identity with women and minorities, multiple
research projects have found.

"Anytime they pulled anyone over, they wanted to run a face scan" against a
database of suspects, Smith said without naming the agency.  After thinking
through the uneven impact, "we said this technology is not your answer."

Speaking at a Stanford University conference on "human-centered artificial
intelligence," Smith said Microsoft had also declined a deal to install
facial recognition on cameras blanketing the capital city of an unnamed
country that the nonprofit Freedom House had deemed not free.  Smith said it
would have suppressed freedom of assembly there.

On the other hand, Microsoft did agree to provide the technology to an
American prison, after the company concluded that the environment would be
limited and that it would improve safety inside the unnamed institution.

Smith explained the decisions as part of a commitment to human rights that
he said was increasingly critical as rapid technological advances empower
governments to conduct blanket surveillance, deploy autonomous weapons and
take other steps that might prove impossible to reverse.

Microsoft said in December it would be open about shortcomings in its facial
recognition and asked customers to be transparent about how they intended to
use it, while stopping short of ruling out sales to police.

Smith has called for greater regulation of facial recognition and other uses
of artificial intelligence, and he warned Tuesday that without that,
companies amassing the most data might win the race to develop the best AI
in a "race to the bottom."

He shared the stage with the United Nations High Commissioner for Human
Rights, Michelle Bachelet, who urged tech companies to refrain from building
new tools without weighing their impact.

"Please embody the human rights approach when you are developing
technology," said Bachelet, a former president of Chile.

Microsoft spokesman Frank Shaw declined to name the prospective customers
the company turned down.

Reporting by Joseph Menn; Editing by Greg Mitchell and Lisa Shumaker

Frank Konkel, 17 Apr 2019

Microsoft Unveils Two Secret Data Centers Built for Classified Government

... Microsoft's announcement is part of the company's plan to compete with
Amazon--the only company cleared to host the CIA and Defense Department's
secret and top secret classified data--and comes as both companies compete
for a $10 billion military cloud contract called *JEDI*.  ...

Microsoft Email Hack Shows the Lurking Danger of Customer Support (WiReD)

Gabe Goldberg <>
Tue, 16 Apr 2019 20:22:03 -0400
On Friday night, Microsoft sent notification emails to an unknown number of
its individual email users—across Outlook, MSN, and Hotmail—warning
them about a data breach. Between January 1 and March 28 of this year,
hackers used a set of stolen credentials for a Microsoft customer support
platform to access account data like email addresses in messages, message
subject lines, and folder names inside accounts. By Sunday, it acknowledged
that the problem was actually much worse.

After tech news site Motherboard showed Microsoft evidence from a source
that the scope of the incident was more extensive, the company revised its
initial statement, saying instead that for about 6 percent of users who
received a notification, hackers could also access the text of their
messages and any attachments. Microsoft had previously denied to TechCrunch
that full email messages were affected.

As China Hacked, U.S. Businesses Turned A Blind Eye (

Richard Stein <>
Wed, 17 Apr 2019 12:33:07 +0800

"Technology theft and other unfair business practices originating from China
are costing the American economy more than $57 billion a year, White House
officials believe, and they expect that figure to grow.

"Yet an investigation by NPR and the PBS television show Frontline into why
three successive administrations failed to stop cyberhacking from China
found an unlikely obstacle for the government—the victims themselves."

Why do for-profit organizations, possessing vast stores of valuable
intellectual property, apparently accept and anticipate theft of this
content?  Because the PRC marketplace is "too big" to ignore.

US businesses display a remarkable, and convenient, myopia when it suits
their primary objective: capture and realize revenue. Corporations are
inured to theft and breach, exhausted by defense against the inevitable.

Businesses budget for theft losses and pay insurance premiums as an
operational expense. No longer is an eyelash of concern raised. These
expenses are considered leakage. (See the movie classic "Casino.").
Business continuity is the objective.

When pushed against the wall (if revenue capture is threatened by
'unfavorable or unfair' competition), business can prevail upon political
governance to embargo foreign-products, or savage their competitor's product
capabilities like HuaWei 5G per

A calculated brand outrage assault and reputation sabotage campaign can tip
procurement scales against certain suppliers.

Given visible product defect escape and zero-day density reports (as noted
in RISKS-31.16 and elsewhere), how do data breach and IP theft incidents
arising from deployed gear (be they domestic or foreign), constitute a
favorable outcome for dependent end-users and businesses?

Whether the PRC or the US/EU "wins the contest" for most rapacious and
effective data breach and IP theft exploitation capabilities is immaterial
to governments.

International economic dominance—hegemony—appears to motivate PRC IP
theft and intrusion frequency: Become the world's largest economy and bask
in the bragging rights limelight by any conceivable means. The US/EU
apparently do not enlist their intelligence services for this purpose, at
least as vigorously engaged or as visibly compared to the #2 global economy.

Risks: Exhausted business strategies and weak operational practices that
rely on government intervention to rebalance the marketplace.  Insufficient
or ineffective safeguards applied to suppress IP Internet theft, intrusions,
and digital data exfiltration.

Wipro customers hacked, says Krebs. Nothing to see here, says Wipro (TechBeacon)

Gabe Goldberg <>
Thu, 18 Apr 2019 13:38:48 -0400

Facebook has admitted to unintentionally uploading the address books of 1.5 million users without consent (The Guardian)

the keyboard of geoff goodfellow <>
Thu, 18 Apr 2019 08:05:53 -1000

Facebook has admitted to `unintentionally' uploading the address books of
1.5 million users without consent, and says it will delete the collected
data and notify those affected.

The discovery follows criticism of Facebook by security experts for a
feature that asked new users for their email password as part of the sign-up
process. As well as exposing users to potential security breaches, those who
provided passwords found that, immediately after their email was verified,
the site began importing contacts without asking for permission.

Facebook has now admitted it was wrong to do so, and said the upload was
inadvertent.  “Last month we stopped offering email password verification
as an option for people verifying their account when signing up for Facebook
for the first time,'' the company said.  “When we looked into the steps
people were going through to verify their accounts we found that in some
cases people's email contacts were also unintentionally uploaded to Facebook
when they created their account, We estimate that up to 1.5 million people's
email contacts may have been uploaded. These contacts were not shared with
anyone and we're deleting them. We've fixed the underlying issue and are
notifying people whose contacts were imported. People can also review and
manage the contacts they share with Facebook in their settings.''

The issue was first noticed in early April, when the Daily Beast reported
on Facebook's practice of asking for email passwords to verify new users. The
feature, which allows Facebook to automatically log in to a webmail account
to effectively click the link on an email verification itself, was
apparently intended to smooth the workflow for signing up for a new account.

But security experts said the practice was `beyond sketchy', noting that
it gave Facebook access to a large amount of personal data and may have led
to users adopting unsafe practices around password confidentiality. The
company was “practically fishing for passwords you are not supposed to
know,'' according to cybersecurity tweeter e-sushi who first raised concern
about the feature, which Facebook says has existed since 2016...

Utah Bans Police From Searching Digital Data Without A Warrant, Closes Fourth Amendment Loophole (Forbes)

Monty Solomon <>
Thu, 18 Apr 2019 11:00:29 -0400

AppleWatch or AnkleMonitor: You Decide

Henry Baker <>
Fri, 12 Apr 2019 07:01:53 -0700
"Ankle monitor" and Fitbit/AppleWatch are becoming indistinguishable in the
new world of Chinese/Uber/AirBnB-style Social Credit Systems.

Three excellent 11-16 minute videos of Big Tech's version of Social
Credit Systems in action.  Well done, with high production values.

This dystopian world is no longer "far into the future", but already



Published on Apr 9, 2019

Jai's celebratory social media post affects her access to vital medication.
Her attempts to circumvent the system leads to even more dire consequences.

Written by: Nehal El-Hadi  Directed by: Josh Lyon


Published on Apr 9, 2019

A smart city tracks and analyzes a woman walking through the city.
Things she does are interpreted and logged by the city system, but are
they drawing an accurate picture of the woman?

Written by: Madeline Ashby   Directed by: Farhad Pakdel

A Model Employee

Published on Mar 29, 2019

To keep her day job at a local restaurant, Neeta, an aspiring DJ, has
to wear a tracking wristband.  As it tracks her life outside of work,
she tries to fool the system, but a new device upgrade means trouble.

Written by: Tim Maughan   Directed by: Leila Khalilzadeh

Fintech fiddles as home burns: 97% of apps lack basic security (TechBeacon)

Gabe Goldberg <>
Fri, 12 Apr 2019 18:46:56 -0400
This is not fine. A white-hat researcher examined 30 financial apps, looking
for information security issues—worryingly, all but one of them were

The failures were mind-numbingly familiar, and dead easy to find. It's as if
the industry has learned nothing and is walking around with a sign on its
back, saying, "Rob me."

Please report problems with the web pages to the maintainer