The RISKS Digest
Volume 32 Issue 70

Saturday, 5th June 2021

Forum on Risks to the Public in Computers and Related Systems

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

Please try the URL privacy information feature enabled by clicking the flashlight icon above. This will reveal two icons after each link the body of the digest. The shield takes you to a breakdown of Terms of Service for the site - however only a small number of sites are covered at the moment. The flashlight take you to an analysis of the various trackers etc. that the linked site delivers. Please let the website maintainer know if you find this useful or not. As a RISKS reader, you will probably not be surprised by what is revealed…

Contents

WARNING to RISKS readers
PGN
Tesla activates in-car camera to monitor drivers using Autopilot
TechCrunch
Tesla brings the strategies pioneered by Apple to the auto industry
WashPost
Tesla apologizes after man in S.China locked in his car due to power failure
Global Times
A “lethal” weaponized drone “hunted down a human target” without being told to for the first time
Business Insider
AI in medicine
Statnews via Wendy Grossman
AI Drone May Have Acted on Its Own in Attacking Fighters, U.N. Says
NYTimes
Don't End Up on This Artificial Intelligence Hall of Shame
WiReD
Bug in Siemens PLCs…
The Hacker News bia Robert Mathews
Cyberattack closes JBS meat-packing facilities in Canada, U.S. and Australia
CBC
How to Negotiate with Ransomware Hackers
The New Yorker
Malware Can Use This Trick to Bypass Ransomware Defense in Antivirus Solutions
The Hacker News
This $5 billion insurance company likes to talk up its AI. Now it's in a mess over it.
cnn.com
Steamship authority targeted in ransomware attack
The Martha's Vineyard Times
Cybersecurity insurance, if you can get it
knowbe4
Supreme Court narrows cybercrime law
The Hill
High-tech policing: Suspect identified after posting pic of his hand holding cheese
LinkedIn
Our digital pasts weren't supposed to be weaponized like this
NYTimes
Will the Excelsior Pass, New York's Vaccine Passport, CatchOn?
NYTimes
How do you know this isn't a fake posting?
Rob Slade
Amazon “stealing” your data is not the same as what Comcast is doingxo
Lauren Weinstein
Amazon Sidewalk Poised to Sweep You Into Its Mesh
ThreatPost
Emergency Amazon
Rob Slade
Amazon home devices may now use part of your WAN uplink for a mesh network with neighbors' Amazon Devices
Newser
FCC's emergency connectivity funds ineligible for school and library self-provisioned networks
Broadband Breakfast
E-Commerce liability cases could open floodgates for lawsuites, panelists agree
Broadband Breakfast
Norton Antivirus Is Now a Cryptominer; Wait, what
Review Geek
The Mayor of Reno Is Betting Big on the Blockchain
WiReD
Oximeters used to be designed for equity. What happened?
WiReD
One blessing of the Cybersecurity Executive Order
Hagai Bar-El
CDC loosened mask guidance to encourage vaccination—it failed spectacularly
Beth Mole, Ars Technica
Deter prying eyes by locking your own letters
Atlas Obscura
Facebook systematically censoring “vaccine concerns”, regardless of truthfulness
Project Veritas
Facebook suspends Trump for 2 years in response to Oversight Board ruling
WashPost
Google made it nearly impossible for users to keep their location private
Business Insider
Security Engineering: A Guide to Building Dependable Distributed Systems
Ross Anderson, reviewed by Sven Dietrich
Re: Risks: Colonial Pipeline accused of negligence in proposed class action
John Bechtel
Re: Florida governor signs law to block deplatforming of Florida politicians
San Steingold
Re: Irish Health Service hit by ransomware
Patrick O'Beirne
Re: Why GitHub Refuses to Provide Key Evidence to a Man on Death Row
Stephen E. Bacher
Re: NoScript is immoral?
Eli the Bearded Kaufmann John Levine
Re: Security of the IMPs
Henry Baker
Re: Truth, Lies, and Automation
Toebs Douglass
Info on RISKS (comp.risks)

WARNING to RISKS readers

Peter Neumann <neumann@csl.sri.com>
Sat, 5 Jun 2021 13:12:19 PDT

It is still taking me several hours to get rid of all the detritus in what is being submitted to RISKS. Office 365 is adding over a hundred lines of cruft in headers to each message. All of the encoded characters created by different mail systems have to be dealt with separately. Therefore, as a cruelty to readers instead of cruelty to myself, the next issue will be RAW RECEIVED TEXT. Perhaps I will first remove the Office 365 cruft on most of the messages, but leave them in for the lead message just for kicks. This will save me a few hours, but perhaps give you some ideas of why this is so painful, and how contributors might be able to simplify my efforts with just a little awareness of what is being produced.

Dan Jacobson has kindly offered a bunch of excellent suggestions, only some of which I have been able to adopt.


Tesla activates in-car camera to monitor drivers using Autopilot (TechCrunch)

ACM TechNews <technews-editor@acm.org>
Wed, 2 Jun 2021 12:23:54 -0400 (EDT)

Kirsten Korosec, TechCrunch, 27 May 2021, via ACM TechNews, 2 Jun 2021

Electric-vehicle manufacturer Tesla has turned the in-car camera in its Model 3 and Model Y vehicles into a monitor for when its Autopilot advanced driver assistance system is in use. A Tesla software update specified that the “cabin camera above the rearview mirror can now detect and alert driver inattentiveness while Autopilot is engaged,” and that the system can save or transit information [only] if data sharing is intentionally enabled. Tesla has been criticized for failing to activate its in-vehicle driver monitoring technology amid growing evidence that owners were misusing Autopilot. Jake Fisher (Consumer Reports) said, “If the new system proves effective, it could help prevent distraction and be a major improvement for safety—potentially saving lives.”

https://orange.hosting.lsoft.com/trk/click?ref=znwrbbrs9_6-2b464x22b90ex067267&


Tesla brings the strategies pioneered by Apple to the auto industry (WashPost)

Gabe Goldberg <gabe@gabegold.com>
Mon, 31 May 2021 15:24:56 -0400

Tesla is bringing the strategies pioneered by Apple to the auto industry. Consumers are learning that's not always a good thing.

SAN FRANCISCO—Tesla released its futuristic Full Self-Driving package last year to great fanfare, criticism and the usual stream of video uploads showing off cars that could seemingly drive themselves.

Then something strange happened.

The electric-vehicle giant revoked access for some drivers, it said. Tesla CEO Elon Musk announced on Twitter in March that some users who had received access to the company's most advanced driver-assistance features “did not pay sufficient attention to the road.” Tesla did not say how it made the determination or who among the feature's 2,000 beta testers — who shelled out thousands for the package that Tesla now priced at $10,000 — would lose access. […]

The cars' groundbreaking over-the-air updates mean users can be subject to sudden performance changes if products become out of date — like battery throttling for which Apple has come under fire. Tesla's unique systems have also proved difficult for government authorities investigating crashes to decode, a problem that echoes federal authorities' difficulty unlocking Apple devices. […]

Months after buying a used Tesla Model S for nearly $46,000, Harpreet Singh began to notice the car wouldn't travel far enough on a single charge to cover his work trips frequently stretching more than 200 miles.

Tesla had taken about 40 miles of range off his used Model S, which began with 265 miles, in what Tesla said was an effort to protect the battery. The update also slowed down charging times, Singh said. Tesla ultimately agreed to replace what it later concluded was a faulty battery, but at the expense of what Singh has found is slower acceleration.

After the car and its new battery were working properly, Singh began to dread system updates, because they introduced new problems like the shorter range and decreased charging rates.

Singh said he thinks about it like other tech updates. “I'm so comfortable with Windows 8. … Why do I have to change to Windows 10? And then everything breaks,” said Singh, 33, of Cypress, Tex. “Same thing here. … They can do anything to do it.” […]

Full self-driving features are also not transferrable between cars, meaning an owner who has shelled out $10,000 for the software would have to buy it for their next Tesla as well.

Musk has said, however, that Tesla will look into upping the trade-in value for a vehicle with Full Self-Driving, after some owners complained about having to purchase it twice.

https://www.washingtonpost.com/technology/2021/05/14/tesla-apple-tech/


Tesla apologizes after man in S.China locked in his car due to power failure (Global Times)

geoff goodfellow <geoff@iconia.com>
Fri, 4 Jun 2021 13:18:36 -1000

https://www.globaltimes.cn/page/202106/1225359.shtml


A “lethal” weaponized drone “hunted down a human target” without being told to for the first time (Business Insider)

Paul Davey <pd@pdc.co.uk>
May 31, 2021 6:17:29 JST

https://www.businessinsider.com/killer-drone-hunted-down-human-target-without-being-told-un-2021-5?r=US&IR=T

A “lethal” weaponized drone “hunted down a human target” without being told to for the first time, according to a UN report seen by the New Scientist.

The March 2020 incident saw a KARGU-2 quadcopter autonomously attack a human during a conflict between Libyan government forces and a breakaway military faction, led by the Libyan National Army's Khalifa Haftar, the Daily Star reported.

The Turkish-built KARGU-2, a deadly attack drone designed for asymmetric warfare and anti-terrorist operations, targeted one of Haftar's soldiers while he tried to retreat, according to the paper.


AI in medicine (Statnews)

“Wendy M. Grossman” <wendyg@pelicancrossing.net>
Wed, 2 Jun 2021 11:11:57 +0100

Statnews reports on a study of 400 AI models proposed during the pandemic for spotting illness and predicting which patients are most likely to have serious illness…and finds that all of them are flawed in surprisingly obvious ways. Underlying problems of methodology is the paucity of large, available, diverse data sets. https://www.statnews.com/2021/06/02/machine-learning-ai-methodology-research-flaws/

The great thing about machine learning is it does RISKS at scale. WG


AI Drone May Have Acted on Its Own in Attacking Fighters, U.N. Says (NYTimes)

Gabe Goldberg <gabe@gabegold.com>
Sat, 5 Jun 2021 17:45:25 -0400

A United Nations report suggested that a drone, used against militia fighters in Libya's civil war, may have selected a target autonomously.

https://www.nytimes.com/2021/06/03/world/africa/libya-drone.html


Don't End Up on This Artificial Intelligence Hall of Shame (WiReD)

Gabe Goldberg <gabe@gabegold.com>
Thu, 3 Jun 2021 19:46:43 -0400

A list of incidents that caused, or nearly caused, harm aims to prompt developers to think more carefully about the tech they create.

https://www.wired.com/story/artificial-intelligence-hall-shame/


Bug in Siemens PLCs…

“Robert Mathews (OSIA)” <mathews@hawaii.edu>
Fri, 04 Jun 2021 12:57:00 -0700

“A New Bug in Siemens PLCs Could Let Hackers Run Malicious Code Remotely“ Ravie Lakshmanan, The Hacker News, 31 May 2021

https://thehackernews.com/2021/05/a-new-bug-in-siemens-plcs-could-let.html


Cyberattack closes JBS meat-packing facilities in Canada, U.S. and Australia (CBC)

“Matthew Kruk” <mkrukg@gmail.com>
Wed, 2 Jun 2021 06:45:23 -0600

https://www.cbc.ca/news/business/jbs-meat-cyberattack-1.6048942

A ransomware attack against Brazilian meat-packing giant JBS has disrupted production in the U.S., Canada and Australia. JBS is the world's largest meatpacker and the attack caused its Australian operations to shut down on Monday and stopped livestock slaughter at its plants in several U.S. states and the company's facility near Brooks, Alta.

The ransomware attack follows one last month on Colonial Pipeline, the largest fuel pipeline in the U.S., which crippled fuel delivery for several days in the southeastern part of the country.


How to Negotiate with Ransomware Hackers (The New Yorker)

Jan Wolitzky <jan.wolitzky@gmail.com>
Mon, 31 May 2021 12:25:54 -0400

Rachel Monroe, Annals of Technology, 7 Jun 2021

Kurtis Minder finds the cat-and-mouse energy of outsmarting criminal syndicates deeply satisfying, 31 May 2021

https://www.newyorker.com/magazine/2021/06/07/how-to-negotiate-with-ransomware-hackers


Malware Can Use This Trick to Bypass Ransomware Defense in Antivirus Solutions ()

geoff goodfellow <geoff@iconia.com>
Tue, 1 Jun 2021 10:03:23 -1000

Researchers have disclosed significant security weaknesses in popular software applications that could be abused to deactivate their protections and take control of allow-listed applications to perform nefarious operations on behalf of the malware to defeat anti-ransomware defenses.

The twin attacks, detailed <https://dl.acm.org/doi/10.1145/3431286> by academics from the University of Luxembourg and the University of London, are aimed at circumventing the protected folder feature offered by antivirus programs to encrypt files (aka “Cut-and-Mouse”) and disabling their real-time protection by simulating mouse “click” events (aka “Ghost Control”).

“Antivirus software providers always offer high levels of security, and they are an essential element in the everyday struggle against criminals,” said <https://wwwen.uni.lu/university/news/latest_news/researchers_discover_fix_vulnerability_in_antivirus_software> Prof. Gabriele Lenzini, chief scientist at the Interdisciplinary Center for Security, Reliability, and Trust at the University of Luxembourg. “But they are competing with criminals which now have more and more resources, power, and dedication.”

Put differently, shortcomings in malware mitigation software could not just permit unauthorized code to turn off their protection features, design flaws in Protected Folders solution provided by antivirus vendors could be abused by, say, ransomware to change the contents of files using an app that's provisioned write access to the folder and encrypt user data, or a wipeware to irrevocably destroy personal files of victims. […]

https://thehackernews.com/2021/06/malware-can-use-this-trick-to-bypass.html


This $5 billion insurance company likes to talk up its AI. Now it's in a mess over it. (cnn.com)

Richard Stein <rmstein@ieee.org>
Tue, 1 Jun 2021 11:59:46 +0800

https://edition.cnn.com/2021/05/27/tech/lemonade-ai-insurance/index.html

“But in Lemonade's IPO paperwork, filed with the Securities and Exchange Commission last June, the company wrote that AI Jim ‘handles the entire claim through resolution in approximately a third of cases, paying the claimant or declining the claim without human intervention.’”

Lemonade walked-back that statement—post-IPO, and after Twitter blasted the brand for claiming their AI ‘Jim’ dispensed claim adjustment based on facial recognition.

Expect one or more lawsuits from investors who drank the lemonade without reading the label.

Risk: Overtrust reliance on AI business solution capabilities and commercial viability.


Steamship authority targeted in ransomware attack (The Martha's Vineyard Times)

<>
Wed, 2 Jun 2021 15:17:04 PDT From: Peter Neumann <neumann@csl.sri.com>

https://www.mvtimes.com/2021/06/02/ssa-targeted-ransomware-attack/

[Wiped out the ability to run operations online. Long delays.]


Cybersecurity insurance, if you can get it (knowbe4)

Paul Burke <box1320@gmail.com>
Tue, 1 Jun 2021 10:39:32 -0700

Article from IT security consultant, about ransomware insurance: https://blog.knowbe4.com/cybersecurity-insurance-landscape-is-fundamentally-changing-right-now


Supreme Court narrows cybercrime law (The Hill)

Peter Neumann <neumann@csl.sri.com>
Thu, 3 Jun 2021 10:21:14 PDT

Chris Mills Rodrigo, 3 Jun 2021

https://thehill.com/policy/technology/556686-supreme-court-narrows-cybercrime-law

The Supreme Court limited the scope of a crucial federal computer fraud law Thursday by overturning the conviction of a former police officer accused of misusing a government database.

The justices sided 6-3 with Georgia police sergeant Nathan Van Buren in his appeal of a conviction under the Computer Fraud and Abuse Act. Conservative Justices Clarence Thomas, John Roberts and Samuel Alito dissented.

The 1986 law prohibits accessing a computer “without authorization or exceeding authorized access.”

The Justice Department had argued that Van Buren ran afoul of that law when he took a bribe to access a woman's license plate information in what was a 2015 FBI sting operation. The former officer had argued that that interpretation was too broad because he did have legitimate access to the database, even if he misused it.

If simply violating the terms of a system is illegal under the CFAA, his team argued, then people could be charged for things as mundane as using work computers for personal use.

The majority opinion, penned by Amy Coney Barrett, echoed that assessment. “The Government's interpretation of the ‘exceeds authorized access’ clause would attach criminal penalties to a breathtaking amount of commonplace computer activity,” the opinion reads. “For instance, employers commonly state that computers and electronic devices can be used only for business purposes. On the Government's reading, an employee who sends a personal e-mail or reads the news using a work computer has violated the CFAA.”


High-tech policing: Suspect identified after posting pic of his hand holding cheese (LinkedIn)

Gabe Goldberg <gabe@gabegold.com>
Wed, 2 Jun 2021 13:29:52 -0400

https://www.linkedin.com/posts/christian-quinn_innovation-technology-policy-activity-6803646475923443712-H7ph/


Our digital pasts weren't supposed to be weaponized like this (NYTimes)

Gabe Goldberg <gabe@gabegold.com>
Sat, 5 Jun 2021 17:46:21 -0400

A recent firing at The Associated Press is the latest example of the way in which our digital pasts are never far from the present, despite what early internet evangelists thought.

https://www.nytimes.com/2021/05/29/technology/emily-wilder-firing-ap.html


Will the Excelsior Pass, New York's Vaccine Passport, CatchOn?

Monty Solomon <monty@roscom.com>
Tue, 1 Jun 2021 20:26:58 -0400

More than 1 million Excelsior passes have been downloaded since they were introduced, but officials are hoping they will be adopted more widely.

https://www.nytimes.com/2021/06/01/nyregion/excelsior-pass-vaccine.html


How do you know this isn't a fake posting? (Rob Slade)

Rob Slade <rslade@gmail.com>
Tue, 1 Jun 2021 10:19:00 -0700

90% of Americans think they are better than average at detecting em>fake news, Which is impossible, and they aren't as good as they think they are.

https://lite.cnn.com/en/article/h_077b962ec93232039cadc784d15124a5

Krueger-Dunning lives …


Amazon “stealing” your data is not the same as what Comcast is doing

Lauren Weinstein <lauren@vortex.com>
Sun, 30 May 2021 08:42:59 -0700

There is some confusion about what Comcast is doing when it sets up public Wi-Fi using customers' in-home modems, vis-a-vis what Amazon's new data “stealing” scheme is doing. There are big differences.

1) Comcast is setting up essentially a separate virtual LAN for the public Wi-Fi that does not interact with your normal data flows.

2) Comcast is adjusting for that secondary usage so that it has no impact on your usage costs or usable bandwidth.

Amazon is just taking your data without your affirmative permission, to service their other customers.


Amazon Sidewalk Poised to Sweep You Into Its Mesh (ThreatPost)

Monty Solomon <monty@roscom.com>
Fri, 4 Jun 2021 13:59:27 -0400

https://threatpost.com/amazon-sidewalk-to-sweep-you-into-its-mesh/166581/


Emergency Amazon

Rob Slade <rslade@gmail.com>
Tue, 1 Jun 2021 10:43:33 -0700

Amazon is partnering with aid organizations, including the Red Cross, to get disaster relief materials to where they are needed in a disaster. https://lite.cnn.com/en/article/h_bc341a644b497f6388fd9bfdbc8a6db3

On the one hand, it's great to see a giant corporation helping out.

On the other hand, does Amazon become a single point of failure for disaster relief?


Amazon home devices may now use part of your WAN uplink for a mesh network with neighbors' Amazon Devices (Newser)

Bob Gezelter <gezelter@rlgsc.com>
Thu, 3 Jun 2021 07:32:38 -0400

An interesting and unsettling development on multiple levels. First, there is the technical issue of whether the implementation is truly secure, including whether information can be deduced from such activity. Second, there is a question of propriety. Is it desirable for that level of personal observation to be transmitted outside the residence. Thirdly, is taking any amount of my paid for bandwidth legal and acceptable? An additional, and perhaps more important question is whether such a feature should be enabled by default. NOTE: The referenced article contains a number of web references to The Guardian, Ars Technica, and other mainstream sources.

https://www.newser.com/story/306874/amazon-is-about-to-take-and-share-a-slice-of-your-internet.html


FCC's emergency connectivity funds ineligible for school and library self-provisioned networks (Broadband Breakfast)

Gabe Goldberg <gabe@gabegold.com>
Tue, 1 Jun 2021 14:17:53 -0400

But when the rules on how to spend the money were finalized on May 10th, the FCC's Report and Order declared that schools and libraries could not use Connectivity Funds to build self-provisioned networks, but instead could only use the funds to purchase Wi-Fi hotspots, modems, routers, and connected devices, such as laptop computers and tablets.

The one exception in which schools and libraries can use Connectivity Funds to build self-provisioned networks is in “areas where no service is available for purchase,” based on data self-reported by private ISPs.

The Report and Order indicates the agency was not convinced allowing schools and libraries to build their own networks with the funds would be consistent with the goals Congress intended for the program, as the language in the Rescue Plan states that the Connectivity Fund is limited to the purchase of eligible equipment or advanced telecommunications and information services, as defined here.

What's striking about that FCC interpretation is that it is completely at odds with what the Biden Administration has been espousing in the American Jobs Plan: that building publicly-owned community networks and investing in future-proof infrastructure are a crucial part of closing the digital divide. This FCC decision is a recipe for cutting students off from broadband Internet access as soon as Congressional appropriations run out rather than using those funds for solutions that will operate sustainably into the future.

Not Trying to Rock the Big Telco Boat

When the Connectivity Fund was first introduced, smaller Internet Service Providers, public interest groups, and education advocates petitioned the FCC to allow for the federal funds headed to schools and libraries to be eligible for use to build school and community networks.

The Schools, Health and Libraries Broadband Coalition; the American Library Association; and the Consortium for School Networking all found that self-provisioned networks are the most cost-effective way to permanently close the homework gap. They advocated for giving schools and libraries the most flexibility to spend these dollars and maintained that local administrators are best positioned to decide how to bridge gaps in connectivity.

Instead, the Connectivity Fund is now set to give limited remote learning funds to the same corporate ISPs that gave rise to the homework gap in the first place. The program gives a strong preference to funding hotspots provided by existing wireless mobile service providers, mainly AT&T, Verizon, and T-Mobile. (In fact, AT&T, Verizon, and CenturyLink all lobbied the agency to disqualify [pdf] self-provisioning from being eligible for ECF support.)

The agency has also announced that the program will be forward-looking; therefore, lower priority will be placed on reimbursing schools and libraries for equipment purchased over the past year to expand existing networks or build new networks to serve students and library patrons.

https://broadbandbreakfast.com/2021/05/fccs-emergency-connectivity-funds-ineligible-for-school-and-library-self-provisioned-networks/


E-Commerce liability cases could open floodgates for lawsuits, panelists agree (Broadband Breakfast)

Gabe Goldberg <gabe@gabegold.com>
Tue, 1 Jun 2021 14:20:16 -0400

Amazon is entangled in local legal cases that could set off lawsuits for third-party products sold on its platform.

May 27, 2021”Emerging legal rulings holding online retailers liable for defective third-party products could cause a ripple effect of lawsuits if more courts across the nation adopt that position, according to a panel of legal experts at an event hosted by the Information Technology & Innovation Foundation on Wednesday.

Product liability law has traditionally held that the “seller” of products are responsible for the defects those products may have. You buy a curling iron from Target, for instance, not directly from Dyson. Target is the seller, and in the case of product defection, Target may be the responsible party.

But Amazon has avoided the legal distinction of seller until recently by arguing that they merely act as the middleman in transactions, and that when items are purchased from its website, business is done directly with the manufacturer, which would be responsible in any legal proceeding. Some have argued that this insulation from liability has made e-commerce companies like Amazon far too powerful.

But two rulings in California and one outstanding case in Texas are challenging that assumption.

https://broadbandbreakfast.com/2021/05/e-commerce-liability-cases-could-open-floodgates-for-lawsuits-panelists-agree/


Norton Antivirus Is Now a Cryptominer; Wait, what (Review Geek)

Gabe Goldberg <gabe@gabegold.com>
Fri, 4 Jun 2021 18:08:52 -0400

You can't be serious. Norton 360, the somewhat-frustrating antivirus software that comes preinstalled on many Windows computers, will soon have a built-in Ethereum cryptominer. In its press release, NortonLifeLock says that Norton Crypto will empower people to mine with a “brand they trust” instead of taking risks and running “unvetted code” on their computers. […]

But let's be realistic for a second”the kind of people who will use Norton Crypto probably wouldn't go out of their way to download a spooky, “unvetted” cryptomining software. They will only use Norton Crypto because it came preinstalled on their computer and, at a glance, produces free money. Norton Crypto users may not fully understand how the software works, the impact that cryptomining has on their computer's lifespan, the tax requirements for cryptomining, or the risks involved with crypto trading.

At its launch, Norton Crypto will only produce Ethereum, which is difficult to mine on a single laptop or desktop. As noted by the BBC, it looks like NortonLifeLock will get around the problem by combining miners' computing power into a “pool” and divvying up earnings. Problem is, it's common for crypto pools to have a 1% fee. If Norton Crypto relies on such a system, then NortonLifeLock could develop an extremely lucrative revenue stream at the expense of its customers' computer hardware and naïvety.

https://www.reviewgeek.com/86346/norton-antivirus-is-now-a-cryptominer-wait-what/

News that sounds like a joke. I ran Norton SystemWorks, then Norton 360. Gave it up because … I forget why; maybe too heavy a footprint, too expensive, maybe Windows Defender and such became good enough. I've never missed it or Norton itself.


The Mayor of Reno Is Betting Big on the Blockchain (WiReD)

Gabe Goldberg <gabe@gabegold.com>
Wed, 2 Jun 2021 00:37:33 -0400

But this spring, [Mayor] Schieve (pronounced SHE-vee) devised a potential solution: a non-fungible token, or NFT, offered for sale on a blockchain called Tezos. The new owner would receive a .CAD file and a video from the artist, but the actual, physical sculpture would stay in that downtown Reno plaza. The proceeds would raise funds for the city to clean up the whale and preserve it for the public to enjoy. Schieve realized this type of semi-symbolic sale might require some sweetening. So she was contemplating offering benefits, like tagging along on her annual trip to Burning Man with fellow elected officials. (They don't stay overnight, Schieve adds; she did not intend to jeopardize any future electoral campaigns with drugs and orgies.)

The issuance of an NFT is not, at this point, such a radical thing, even for a government. Cities and states all over have sought at times to forge links to the blockchain. In 2018, Cleveland declared itself Blockland, though the label seems to have waned. Wyoming has set itself up as the premier regulatory haven for cryptocurrency, a label that other states, including Nevada, now seek to challenge. All it takes is a few interested businesspeople and elected officials receptive to “new ideas,” especially those with a cypherpunk ring. That's not quite what's happening in Reno. For Schieve, the NFT was a gateway to something else.

An early sign emerged in January, when Mayor Francis Suarez of Miami, a person on a recent tear of throwing out tech-friendly ideas and seeing what sticks, tweeted about turning his city into a “hub for crypto innovation” centered around Bitcoin. Schieve was unsatisfied. “When are you going to become a $LINK marine?” she teased in reply, cryptically to most readers. She was referring to a blockchain platform called Chainlink, perhaps best known for its cult following of “marines” who swarm toward any mention of the technology on social media. Their loyalty is expressed through ranks earned by #HODLing (that is, holding) the platform's cryptocurrency, called Link. Apparently, the mayor of Reno was a member of the battalion—“link pilled,” in the community's parlance. “It was really sweet,” Schieve says of the meme invasion her tweet inspired.

https://www.wired.com/story/mayor-reno-betting-blockchain/


Oximeters used to be designed for equity. What happened? (WiReD)

Gabe Goldberg <gabe@gabegold.com>
Fri, 4 Jun 2021 18:31:20 -0400

The pandemic drew attention to the racial bias built into pulse oxes. But calls to create a fairer device are missing one thing: It once existed.

https://www.wired.com/story/pulse-oximeters-equity/


One blessing of the Cybersecurity Executive Order

Hagai Bar-El <info@hbarel.com>
Sat, 5 Jun 2021 21:18:52 +0300

On May 12th, the Biden administration issued an Executive Order <https://www.whitehouse.gov/briefing-room/presidential-actions/2021/05/12/executive-order-on-improving-the-nations-cybersecurity/> that was written to improve the overall security posture of software products that the government buys from the private sector. Recent events, such as the SolarWinds hack <https://www.crn.com/the-solarwinds-hack>, contributed to the realization that such a move is necessary.

This Executive Order is a big deal. Of course, nothing will change overnight, but given the size and complexity of the software industry, as well as the overall culture behind software security (the culture of: “If the customer doesn't see it — don't spend money on it”), an Executive Order can probably yield the closest thing to immediate improvement that we could reasonably wish for. The US Government is a very large customer, and all major vendors will elect to comply with its requirements rather than cross it all off their addressable markets.

A lot has been written on how important it is for the government to use its buying power (if not its regulatory power) to drive vendors into shipping more secure products. Product security suffers from what could best be described as a /market failure/ condition, which would call for such regulatory intervention.

To not overly repeat the mainstream media, I would like to focus on one unique aspect of the current Executive Order, and on how it can ignite a new trend that will change product and network security for the better. I'll discuss true machine-readable security documentation.

The requirement for a Software Bill of Materials

The Executive Order requires that every software product is accompanied by a Software Bill of Materials (SBOM) which lists the third-party software modules that it contains. This is essential for the customer to monitor its exposure to supply-chain vulnerabilities. Let's say that I ship software /X/, and that my software happens to utilize a library from a third party, say, the library /L/. I now need to worry not only about vulnerabilities found in my software /X/, but also about vulnerabilities found in /L/. If I am careless, I could keep maintaining my own software without incorporating patches that are made available for /L/ by its vendor. If I am negligent, I would even not bother to check if there are any newly discovered vulnerabilities that need to be patched in /L/. If I am yet more negligent, I could even keep on using a stale end-of-life /L/ library that is no longer maintained at all. However careless or negligent I am, the price is always to be paid eventually by my customer. The customer might not even /know/ how reliant his security posture is on /L/. For what he knows, he only bought /X/. If that customer knew that its security posture relies on /L/ as well, it could have put pressure on me to make sure I use a secure version of /L/ in my product, or not use it at all. The customer, in other words, could use its buying power to improve what it gets. This

situation is what the SBOM provision comes to solve. It forces me to disclose to my customers those additional dependencies that I subject them to, so they can exert their market power towards improving the quality of what they get.

The bigger picture: security documentation

The SBOM is a great idea, and its benefit is yet wider. Security documentation is in poor shape today. Security is not very well covered in product documentation. Technical specifications, like the RFCs published by the Internet Engineering Task Force (IETF), have sections titled Security Considerations in them; product documentation usually doesn't. Even answers to basic questions such as: “What are the exposure risks to the data processed by the product?” or “What could I do as a customer to minimize the attack surface of the product I'm using?” are seldom answered directly. If the customer is lucky, then there are tips scattered around the manuals, help pages, and readme files. If the customer is less lucky, as customers usually turn out to be in such cases, he will need to deduce this from other pieces of product information.

Any security-specific documentation that products will now have to be shipped with is an immense improvement, and will hopefully serve as a precedent for more. One day, hopefully, customers will require a clear manifest of the product's attack surface: an enumeration of all interfaces and how those are protected. Cynicism aside, I am confident that once vendors actually produce such documents for their customers, they may become aware of some vulnerabilities of their products of which they were not aware before, and fix them on time.

Once we generate more security documents for products, the next step would be making those security documents truly machine-readable.

True machine-readable security documents

The Executive Order requires the SBOM document to be included with the product, without prescribing the precise format this document should take, but noting that it shall be ‘machine-readable’. Every vendor can use whatever format it desires, and ‘machine-readable’ is a definition that is wide enough to cover any document which is not a handwritten napkin (until it gets scanned). Nevertheless, we are likely to witness accelerated document evolution. Thousands of vendors will have to start producing those documents very shortly. It will take very few years, rather than decades, for the industry to converge onto a few stable forms (most likely the forms that will be used by the major consultancies and certification bodies, and in light of further instructions from the government). The standardization fora will soon enough take on the challenge of defining a standard schema, augmenting some work that has already been done.

When this happens, we will all be one large step into the future of true machine-readable security documentation. By ‘/true/ machine-readable’ I refer to documents that machines can actually learn from, not just parse.

Once the SBOM document uses a true machine-readable format, it will be processed by risk management software packages. Such packages will take this input, along with assessment and prioritization from tools like /Kenna/ or /VulnDB/, to draw a more accurate risk posture for the organization, based on the newly learned dependencies. Introducing automation into the process will also force the vendors into keeping those SBOM documents accurate and updated.

The prevalence of security documents that are truly machine-readable is a big deal. We are not just talking about a security document that is read by a management app instead of by a person; we are talking about a step in the direction of reducing one of the biggest headaches of security monitoring configuration: discovery.

A headache called discovery

The year is 1997, and I get to help improve the security of a large organization. One big challenge at the time was the connection of desktops using modems that were left in answer mode when unattended. I came prepared with instructions and scripts for securing those modems. Soon enough I learned that there was no place in the organization where all those modems were even registered. The one-month “secure the modems” project started with 3.5 weeks of running war-dialers — bots that dial all extensions to create the list of active modems, with just one short week left for actually securing them. Today we barely use modems, but corporate networks grow faster than anyone can keep record, and the trend (at least in tech companies) is to not restrict adoption of new technologies by people, unless necessary. Be it software packages, web services, connected devices or modems, discovery is always a challenge, and the place where many balls get dropped.

Much of the unaddressed attack surface in large systems is caused not by vulnerabilities of which you are unaware, but rather by functionality of which you are unaware. (No point Googling it; I made it up.)

Having mechanisms in place that enforce rigorous record-keeping of systems and their dependencies might not count as the latest core security tech, but can certainly prevent many security incidents.

Beyond SBOM

Once we get into the habit of deploying systems that come with written manifests of their capabilities, there is no reason to stop at the SBOM. Some people suggest an intuitive extension into what they call a Bill of Behaviors, and one can easily think of other security-related properties that vendors could report about their systems. So much heuristics are used by security monitoring tools just because there is no clear statement of what an expected behavior of a system is. Using such heuristics not only implies missing alerts, but it also costs us in reduced sensitivity. Heuristics-based security monitors are configured for reduced sensitivity to overcome false-alerts; false-alerts that could easily kill any deployment of a security monitoring tool. Anyone deploying security monitoring tools will tell you that the Achilles heel of those tools is not in the quality of their monitoring technology, but in the complexity of the configuration management that is required to deploy them effectively. By targeting this complexity, we strengthen the weakest link.

Once true machine-readable security docs appear, and some standard for them emerges, security monitoring systems will happily start reading them. We will enjoy less heuristics involved in assessing what packages an installed piece of software /may/ contain within it, or what network traffic is /reasonable/ to see. Finally, once the overall security posture of a system is more deterministic and less reliant on heuristics, there will be an incentive for vendors to exceed the requirements of Executive Orders, and provide more such machine-readable manifests. This will assure them that their systems are not generating false alarms by security monitoring tools.

What about IoT?

So far, we've discussed typical corporate IT networks. Once the trend of machine-readable security documentation gains traction, it may also be adopted into IoT, where its value will be yet magnified.

In the IoT space, heuristics are more prevalent. It's a relatively new domain where standards are fewer and fragmentation rules. There are good companies out there that built complete business models around trying to identify what's running on an IoT network; even just recognizing what types of devices are involved. Security-wise, the IoT space today is where the IT space was two decades ago, with frequent use of weak authentication, use of old software stacks, and over-reliance on obscurity.

Clarity is a good friend of Security, and IoT networks could use much of it.

Summary: the role of the government

The space of IT security, for both corporate networks, home networks and IoT, leaves much to wish for. The market is motivated by functional features, with security taking the back seat. This is the case, to a large extent, because security is evident neither in its existence nor in its absence; a situation that is likely to prevail.

Moreover, product security suffers from significant information asymmetry. The vendor knows much more about the security of its product than the customer (even if such knowledge means knowing that he doesn't really know, as is the case with many vendors). This asymmetry implies that customers cannot properly factor security into their buying decisions, diminishing the ability of the market to fuel improvements, as it does in other areas.

Such conditions, like the related public safety conditions, call for government intervention. In some cases this happens through regulation (e.g., with car seat belts). In softer cases, where life and death do not seem to be directly at stake, the government can still catalyze improvement by using its buying power. In our case, the primary interest of the government might be to protect itself, rather than the public, but the outcome is the same. (It is reasonable to expect that some of the benefit of that buying power, which the taxpayer enables, benefits back the taxpayer, so all is well.)

Forcing software products to come with a Bill of Materials is just part of the benefit of the Executive Order, but I argue that even this addition alone, once imposed on many large vendors, can ignite a multi-phase process of improvement:

  1. starting with one mandatory machine-readable SBOM document that has to be kept up to date,
  2. leading to machine-generated machine-readable documents, which are easier for the vendor to produce and refresh,
  3. to standards-based true machine-readable documents that can be used by security gear, both for SBOM and for other areas where positive attestation by the vendor can help reduce the use of heuristics in security monitoring,
  4. to an overall higher security posture by improving the accuracy and benefit of security monitoring and enforcement tools; accuracy that will also favor vendors.

We do not need an Executive Order for this, but we do need an Executive Order to build the critical mass of demand for machine-readable security documentation that will ignite this entire process.

Whatever the overall aspiration of the government is — I believe that it will get more than it bargained for.

/This essay has also been published at: https://www.hbarel.com/


CDC loosened mask guidance to encourage vaccination—it failed spectacularly (Beth Mole, Ars Technica)

Dewayne Hendricks <dewayne@warpspeed.com>
May 29, 2021 14:31:25 JST

FDA approval and paid time off would make people more likely to get a shot, poll finds. By Beth Mole May 28 2021 <https://arstechnica.com/science/2021/05/cdc-loosened-mask-guidance-to-encourage-vaccination-it-failed-spectacularly/>

The Centers for Disease Control and Prevention stunned health officials and experts on May 13 with the abrupt announcement that people fully vaccinated against COVID-19 could forgo masking in most settings—indoor, outdoor, uncrowded, and crowded alike. The guidance was a stark reversal from the health agency's previous stance, issued just two weeks earlier, that still recommended vaccinated people wear masks among crowds and in many indoor, uncrowded settings.

The CDC said at the time that it was merely following the science for masking. The agency and its director, Rochelle Walensky, highlighted fresh, real-world studies demonstrating COVID-19 vaccines' high efficacy and ability to lower transmission risks. But the update was also part of an overt effort to encourage vaccination among the vaccine hesitant by emphasizing the perks of being vaccinated—like not needing to wear masks anymore and reclaiming other bits of normal life.

That messaging shift came as states across the country started to see their pace of vaccination slow despite a glut of vaccine doses. Numerous polls have indicated that most of the people eager to get vaccinated already have. Now, with just 62 percent of the US adult population vaccinated, much of the remaining unvaccinated portion is either hesitant or resistant to being vaccinated. It's that group of people the CDC was trying to reach with the new mask guidance.

“The science is also very clear about unvaccinated people,” Walensky said during the May 13 press briefing, in which she announced the mask guidance update. “[Unvaccinated people] remain at risk of mild or severe illness, of death, or spreading the disease to others. You should still mask, and you should get vaccinated right away. Your health and how soon you return to normal life before the pandemic are in your very capable hands.”

Mask blunder

The mask update immediately generated confusion and controversy given the reversal and its abruptness. And according to fresh polling data, the guidance failed spectacularly at convincing unvaccinated people to get vaccinated.

In new results from the Kaiser Family Foundation's ongoing COVID-19 vaccine monitoring poll, 85 percent of unvaccinated people said the CDC's loosened mask guidance for fully vaccinated people made no difference to their vaccination plans. Only 10 percent said the change made them more likely to get vaccinated and a final 4 percent or so said the change made them less likely to get a shot. It gets worse. The poll broke unvaccinated people into three groups: people who said they would &definitely not get vaccinated, people who would get vaccinated only if required, or people who would wait and see. Those most resistant to getting vaccinated were the least likely to be swayed by the CDC's latest guidance. Among the definitely not group, 98 percent said the change made no difference to them and the remaining 2 percent said they were less likely to get vaccinated — zero percent said they were more likely to get a vaccine. For the only if required group, 89 percent said the CDC change made no difference.

Overall in the poll—which collects data on a nationally representative sampling of adults—62 percent said they had already gotten their vaccine (which tracks with CDC vaccination data), 12 percent said they would wait and see about vaccination, 7 percent said they would only get vaccinated if they were required, and 13 percent said they would definitely not get vaccinated. That definitely not portion has largely remained the same throughout the polling, which stretches back to December.

While the CDC's loosened masking guidance was clearly not persuasive to the unvaccinated, the poll explored other tactics that could boost vaccination. The two ideas that seemed to have the most sway were: 1) if the Food and Drug Administration grants a vaccine full approval, rather than the current Emergency Use Authorizations (EUA); and 2) if employers provided paid time off to get vaccinated and recover from any side effects, like feeling under the weather the day after a dose.

FDA approval and PTO

A total of 32 percent of unvaccinated people said a full FDA approval (a Biologics License Application [BLA] approval) would make them more likely to get a COVID-19 vaccine. Currently, all three vaccines available in the US have been granted an EUA. The FDA grants EUAs only during public health emergencies, like the COVID-19 pandemic, through a process that is fast-tracked compared with a full BLA approval.

Importantly, both tracks require efficacy and safety data from massive Phase III clinical trials. The main difference between an EUA and full approval is the amount of time that people in the clinical trials are followed after full vaccination. Typically, the FDA likes to have at least six months of follow-up data from a vaccine trail. This allows the trial runners and the FDA to look at how well vaccine protection holds up over that time and if any rare side effects crop up. For an EUA, the follow-up period may only be around two months.

However, the difference is largely moot at this point. With nearly 167 million people in the US alone already given at least one shot, regulators have a wealth of post-market safety data. Also, Pfizer and BioNTech announced in April that they had six-months of trial follow-up data that confirmed the vaccine's high efficacy and found no safety concerns. Earlier this month, Pfizer and BioNTech, as well as Moderna, announced that they havestarted a rolling data-submission process for a BLA. […]


Deter prying eyes by locking your own letters (Atlas Obscura)

Gabe Goldberg <gabe@gabegold.com>
Fri, 4 Jun 2021 14:59:53 -0400

A how-to for those who want to use folds, tucks, slits, and more to turn letters into little works of art.

https://www.atlasobscura.com/articles/letterlocking-how-to

DIY encryption.


Facebook systematically censoring “vaccine concerns”, regardless of truthfulness (Project Veritas)

Marco <listaddr@gmail.com>
Mon, 31 May 2021 10:55:10 +0200

This is not “fighting fake news”, this is pure censorship.

https://www.projectveritas.com/news/breaking-facebook-whistleblowers-expose-leaked-internal-docs-detailing-new/


Facebook suspends Trump for 2 years in response to Oversight Board ruling (WashPost)

Gabe Goldberg <gabe@gabegold.com>
Fri, 4 Jun 2021 15:06:50 -0400

The change is part of a series of responses to the Facebook Oversight Board's ruling on former President Trump. https://www.washingtonpost.com/technology/2021/06/03/trump-facebook-oversight-board/

Risks? Facebook, Trump…


Google made it nearly impossible for users to keep their location private (Business Insider)

Marco <listaddr@gmail.com>
Mon, 31 May 2021 14:38:38 +0200

Google continued collecting location data even when users turned off various location-sharing settings, made popular privacy settings harder to find, and even pressured LG and other phone makers into hiding settings precisely because users liked them, according to the documents.

https://www.businessinsider.com/unredacted-google-lawsuit-docs-detail-efforts-to-collect-user-location-2021-5


Security Engineering: A Guide to Building Dependable Distributed Systems (Ross Anderson)

“Cipher Editor” <cipher-editor@ieee-security.org>
Fri, 4 Jun 2021 13:28:53 -0600

[Cipher is at http://www.ieee-security.org/cipher.html It is published 6 times per year]

Security Engineering: A Guide to Building Dependable Distributed Systems by Ross Anderson Book Review By Sven Dietrich 5/31/21

Wiley Publishing 2020, ISBN ISBN-13: ISBN: 978-1-119-64278-7 (Hardcover) 1232 pages, Third Edition

We live amid constant reminders in real life about what could have been done better from a computer security perspective. When something goes wrong, we find it is a protocol that is exhibiting an exploitable vulnerability, or a software repository that has been infiltrated with code containing a vulnerability, or a critical infrastructure system held for ransom. One wonders what design principles the system authors and builders had considered to mitigate any compromises or to allow them to continue to function in the presence of those compromises. How can we engineer those solutions, how can we build better systems: more secure, more dependable? One book attempts to provide this background.

At over 1200 pages, Ross Anderson's third edition of ‘Security Engineering: A Guide to Building Dependable Distributed Systems’ is a large update after the first edition in 2001 and the second edition in 2008. This is a comprehensive book on security engineering, providing anywhere from an introduction to the various subfields of computer and network security to considerations necessary to building secure and resilient real-world systems, and all the way to identifying research problems that remain to be addressed for the topics in each chapter.

The book is divided into three parts, with a total of 29 chapters, and contains an extensive bibliography. The first part covers the basics, the second part looks at applications of secure systems, and the third part broadly discusses politics, management, and assurance. Each chapter covers several themed subsections, followed by a chapter summary, a set of research problems, and further reading. The chapters read well and flow easily within themselves as well as from one chapter to the next. While it is a a descriptive treatise, not a rigorous mathematical treatment of the various subjects, nonetheless occasional mathematical formulas or charts will pop up inline to illustrate the broad concepts brought forth and to whet the reader's appetite to seek out the original research paper or other references cited.

The first part spans 8 chapters that quickly set the stage for Ross Anderson's approach to the subject matter: ‘What is Security Engineering?’, ‘Who is the Opponent?’, ‘Psychology and Usability’, ‘Protocols’, ‘Cryptography’, ‘Access Control’, ‘Distributed Systems’, and last but not least ‘Economics’. The reader learns about what it means to deal with adversity in the 2020s, identifying the threat models, the pitfalls, and the consequences of not getting security right. The big impact here is from the author's contribution to the security field, the systems view, the psychology and usability aspects, as well as the economics aspects, topics for which the author has organized (or otherwise contributed to) workshops and conferences.

The second part discusses real-world applications of secure systems, covering many decades of security work, from the early days of ‘Multilevel Security’ and ‘Nuclear Command and Control’, to ‘Advanced Cryptographic Engineering’, ‘Biometrics’ and ‘Tamper Resistance’ as well as Digital Rights Management in ‘Copyright and DRM’, to ‘Network Attack and Defence’, ‘Phones’, ‘Locks and Alarms’, just to mention some of the 16 chapters in here. This part is wrapped up with thoughts on ‘New Directions’ in the field, talking among others about the combination of Machine Learning, Artificial Intelligence and Security and what it means for both attacker and defender sides.

The third part covers politics, management, and assurance in four chapters. Here the reader learns about ‘Surveillance or Privacy’, ‘Secure Systems Development’, ‘Assurance and Sustainability’. Controversial topics of surveillance versus privacy are brought up in the context of political and technological settings that have affected Internet users for many years, including wiretapping and censorship. Risk quantification and DevSecOps are brought into the picture here as well. This part wraps up with ‘Beyond “Computer Says No”’, reminding us what Ross Anderson has told us all along in these chapters: think about the big picture, and how does it fit in?

This is a fantastic book for organizing one's thinking about security engineering and design. The reader how all the facets fit together in the real world through both scientific references and anecdotes from the last few decades. The depth is provided, should the reader care to delve deeper, through an absolutely impressive bibliography of close to 2100 entries. The narrative is easy to follow throughout the book, whether the reader is learning about DDoS attacks (always close to my heart), espionage (Snowden's surveillance revelations, for example), security protocol failures, financial transaction protocols, mobile phone security, electronic voting security (very relevant in the last few years), security printing, covert channels, DNS security, deception, or ransomware, among others.

The breadth of the topics covered provides a good perspective for appreciating the impact that good (secure?) design can have on real-world systems that surround us. That is even more so relevant now that the Internet has invaded, uh, permeated our homes with Internet-of-Things devices that make our lives more Internet-centric with all the advantages and risks that come with it.

The accessible style of this book and, most importantly, the relevant context of the discussed secure systems, make for one pleasurable reading. While it could be considered a very comprehensive introduction to the idea of security engineering, there are enough timely and thought-provoking musings to keep more advanced readers interested in seeking out the scientific articles providing the adequate depth, hindsight, and foresight. This book is a must-have if security engineering is your intended field or connected to your field.

Ross Anderson did a great job of producing the third edition of ‘Security Engineering: A Guide to Building Dependable Distributed Systems’ in 2020, a book intended to last for many years. He is a well-known expert in the security field and this overarching treatise makes for one impressive (and heavy!) book. The book is a welcome addition to my bookshelf, to be used as a reference or even textbook in the years to come.

[Sven Dietrich reviews technology and security books for IEEE Cipher. He welcomes your thoughts at spock at ieee dot org.]

Re: Risks: Colonial Pipeline accused of negligence in proposed class action (Bloomberg Law, RISKS-32.69)

John Bechtel <john@bechtel.me.uk>
Tue, 1 Jun 2021 22:31:13 +0100

The idea that Colonial would shut the pipeline down if it can't measure who is getting what product (as I understand the story) sounds very much like the apocryphal story about telephone exchanges (Central Offices or Switches to some) back in the day: What is the purpose of a telephone exchange? “Why, to make telephone calls, of course!” But that is not the answer. The true answer is: to generate billing records. If the hard disk to which billing records are written is full, should the exchange place calls?


Re: Florida governor signs law to block deplatforming of Florida politicians (The Verge, RISKS-32.69)

Sam Steingold <sds@gnu.org>
Thu, 03 Jun 2021 14:27:41 -0400

I think [they] have reinvented Shadow banning. https://en.wikipedia.org/wiki/Shadow_banning

I find it disappointing that such a terrifying risk to free discourse is being advocated here.


Re: Irish Health Service hit by ransomware (BBC, RISKS-32.68)

“Patrick O'Beirne” <pob@sysmod.com>
Mon, 31 May 2021 09:43:09 +0100

Ongoing disruption and consequences, costs : https://www.irishtimes.com/news/health/hse-cyberattack-has-had-devastating-impact-cancer-services-director-says-1.4576211 “The search for handwritten or printed-out notes can exacerbate delays, causing “a devastating impact on . . . the speed at which we can assess patients”.”

https://www.irishtimes.com/news/health/cyberattack-hse-faces-final-bill-of-at-least-100m-1.4577076

“ The cyberattack on IT systems in the health service will cost it at least €100 million, according to chief executive Paul Reid. This is at the lower end of estimates of the total cost, he indicated, and includes the cost of restoring the network, upgrading systems to Microsoft 365 and the disruption caused to patients.” (From Windows 7)

In other news, citizen contra-attackers: https://www.irishtimes.com/news/crime-and-law/members-of-public-send-messages-to-cyber-gang-that-attacked-hse-1.4575230 “ An online message thread established by the cyber gang that attacked the Health Service Executive has been accessed by a number of unknown people, with gardaí trying to establish who they are and what their motivations are. At least one person who accessed the thread sent sexually explicit and racist comments to the attackers in recent days.”


Re: Why GitHub Refuses to Provide Key Evidence to a Man on Death Row (Gizmodo)

“Stephen E. Bacher” <sebmb1@verizon.net>
Mon, 31 May 2021 12:17:23 -0700

Apart from the social media (Facebook/Twitter/etc.) ramifications, this story evokes another risk: the risk of relying without question on “expert” DNA analysis to prove innocence (or guilt).

Some time ago the public radio program “This American Life” featured an in-depth story which delved anecdotally into the ins and outs of analyzing DNA data; it raised some skepticism, at least in my mind, about the accuracy and reliability of the resulting evidence presented in courtrooms.

This is , to be sure, a journalistic issue at least as much as a legal one.


Re: NoScript is immoral? (Re: Ward, RISKS-32.69)

Eli the Bearded <*@eli.users.panix.com>
Wed, 2 Jun 2021 16:55:23 -0400 (EDT)

The Twitter account Sh_t User Story (name censored for profanity filters) has a wealth of examples of bad technology design many of which would be at home with RISKS. All are presented in the “User Story” format. One relevant to this post:

    As a…
        web user
    I want to…
        whitelist news websites from my ad-blocker plugin
    so that…
        I can take a long break between the first two paragraphs of
        the article, and then be served with a paywall

Link ROT-13rd, again for profanity filters:

uggcf://gjvggre.pbz/FuvgHfreFgbel/fgnghf/1352299991969243138

There are real risk lurking in all of this. Some of them:

  1. People who can pay for news get it, but propaganda remains free.
  2. Ads have become the normalized way of making micropayments on the web, but ads frequently include enough unpleasantness that people take a lot of steps to avoid them (NoScript and ad blockers).
  3. Search engine discoverability is critical for many sites, and search engines don't typically run javascript, so JS disabled access often has to work.

Re: NoScript is immoral? (RISKs 32.69) notsp

<kaufmann@winning.com>
Thu, 3 Jun 2021 23:43:53 +0200

In RISKs 32.69, Martin Ward writes:

>Is it really morally wrong to choose not to execute by default every piece
>of code that is handed to you by any web site that you decide to visit?

Of course not. The way I look at is, it's my computer and my Internet connection, both paid for with my dollars. I have every right to exercise full control over what bits are downloaded with that connection and what happens to them after they arrive on my computer. To argue otherwise is to suggest that it's also morally wrong to leave the room during the commercial breaks in television programs. If there are copyright or other considerations the publisher wishes enforce, then they should be at least nominally negotiated before the content is made available (perhaps even if it's only a “click here to accept our terms” button). I guess we're all still waiting for a viable micropayments system.


Re: NoScript is immoral? (Re: Ward, RISKS-32.69)

“John Levine” <johnl@iecc.com>
30 May 2021 23:17:05 -0400

I wouldn't say it's morally wrong, but as I may have said a few times before, reporters need to eat, so you're definitely a freeloader.


Re: Security of the IMPs (Cosell, RISKS-32.69)

Henry Baker <hbaker1@pipeline.com>
Mon, 31 May 2021 07:38:51 -0700

In an episode of the “The Americans” about deep cover Soviet spies, an ARPAnet IMP makes a brief appearance, as well as a PDP-10 they call “The Beast”.

Here is a still clip from the episode showing the front panel of the IMP.

Don Hopkins, Arpanet Bullshit, 21 Oct 2015 From The Americans, Season 2 Episode 7: Arpanet. https://www.youtube.com/watch?v=hVth6T3gMa0

Re: Truth, Lies, and Automation (RISKS-32.69)

Toebs Douglass <risks@winterflaw.net>
Mon, 31 May 2021 12:58:55 +0200
> Among other achievements, it has drafted an op-ed that was commissioned by
> The Guardian,

So, what happened here is that eight different opt-eds were produced by GPT-3; they were all kept short, and this was deliberate, because one of the fundamental and unsolved issues with artificial text generation is its inability to make sense over longer bodies of text; any given sentence is fine, a couple of sentences usually fine, something longer is problematic - and always will be, I suspect, because you'd need such a vast amount of content, to be able to develop a neural net which has seen enough material on enough subjects to be able to fake it for extended bodies of text, that it is impossible - that much content doesn't actually exist. It's a sort of n^n problem. You end up needing an awful lot more data and computational power just to move ahead a tiny bit.

Of these eight documents, the editors at the Guardian then edited them all, as they saw fit, to produce the single document which was published.

I may be wrong, but I suspect they took the most sane paragraphs from the eight attempts, fixed them up, and re-ordered them to make sense.

If you're thinking this whole piece is the direct product of a text generator, it really isn't, and the areas where humans helped are exactly the areas where the method used is fundamentally and inherently weak.

> written news stories that a majority of readers thought were written by
> humans,

This claim is backed up by a link to an arxiv white paper.

In the white paper, various AI models (of increasing size, culminating in GPT-3) were given an original 200 or so word news piece written by a human and asked to generate text based on this primer. The generated text was presented to the humans, who had to decide if it was human or AI written.

I may well just not be seeing it, but all I can see is the claim that as the size of the model increases, the time taken to decide increases, and the success rate drops. No actual numbers appear to be given.

As before, short text is being used because of the fundamental and inherent difficulty in producing longer texts.

> and devised new Internet memes.

This claim is backed up by a link to a tweet. The tweet appears to show in a video of sequential still images a series of short, one or two word phases, submitted to GPT-3 by some guy, and its response. The only other information about what was done is that “explaining the meme in the priming improves the consistency/quality”. Presumably also these represent the best results found, as selected by a human.

> In light of this breakthrough, we consider a simple but important question:
> can automation generate content for disinformation campaigns?

Examining the claims made so far, there has been no breakthrough.

I've not read the document published by the Center for Security and Emerging Technology. It may be it is a well-balanced, rational and reasonable document. However, this one paragraph, being more closely examined, appears to be sensationalism; the claims made are misleading, and seem far in excess of the basis upon which they are made.

Please report problems with the web pages to the maintainer

x
Top