Monday, April 22, 2019


Wouldn’t it be interesting if the hackers had the best AI?
Internet bots are getting smarter and imitating people
Bots are becoming a bigger threat as they get smarter, a new report says.
A whopping 73.6 percent of bad bots are so-called Advanced Persistent Bots, which use anonymous proxies, change their identities and mimic human behavior, according to the 2019 Bad Bot Report from Distil Networks, which provides Web content protection services.
… Sophisticated bad bots reproduce mouse movements and clicks that fool even sophisticated detection methods and mask their activity by “reverse engineering detection systems,” Distil said.
“Advanced attackers now show definitive behavior that they know about the technology they’re trying to defeat, and they’re continuously learning how to adapt their tactics,” the study said.




Short answer: Nope.
IAPP FAQs: Are GDPR-compliant companies prepared for CCPA?
The California Consumer Privacy Act is top of mind for many privacy professionals across the U.S., who are working to leverage their GDPR preparation to build CCPA-compliance programs. They are learning that while their recent GDPR preparation is helpful, the CCPA has nuanced requirements that go beyond the GDPR. Emphasis is often placed on the novel “Do Not Sell My Personal Information” link.
After listening to two useful web conferences comparing the CCPA and GDPR (available here and here, in case you missed them), I wondered if companies outside of the U.S. have realized that California-specific adjustments will be needed. If not, the CCPA’s private right of action and what many consider the litigious nature of the U.S. might soon draw the attention of foreign C-suites.




Does this mean TSA is unable to read the dates on a visa to determine if the passenger has overstayed?
The US wants to scan the faces of all air passengers leaving the country
The US immigration system was designed to track who comes into the country, not who leaves. For more than two decades, authorities have been trying to find an effective way to keep tabs on departing foreigners—and those who overstay their visas.
US Customs and Border Protection (CBP) now says it’s found a solution: facial recognition. It expects to be able to scan 97% of commercial passengers within the next four years, according to a report released by the Department of Homeland Security today.




What employee privacy?
Employee privacy in the US is at stake as corporate surveillance technology monitors workers’ every move
A 2018 survey by Gartner found that 22% of organizations worldwide are using employee-movement data, 17% are monitoring work-computer-usage data, and 16% are using Microsoft Outlook- or calendar-usage data.
… “Employees are in a difficult position. As more and more consumer privacy laws take shape, we’ve seen that there’s been a concern from companies that those privacy laws don’t apply to employees,” said Lee Tien, senior staff attorney at the Electronic Frontier Foundation.
In the workplace, almost any consumer privacy law can be waived. Even if companies give employees a choice about whether or not they want to participate, it’s not hard to force employees to agree. That is, unless lawmakers introduce laws that explicitly state a company can’t make workers agree to a technology, he said.
One example: Companies are increasingly interested in employee social media posts out of concern that employee posts could reflect poorly on the company. A teacher’s aide in Michigan was suspended in 2012 after refusing to share her Facebook page with the school’s superintendent following complaints about a photo she had posted. Since then, dozens of similar cases prompted lawmakers to take action. More than 16 states have passed social media protections for individuals.
… Still, the dearth of specific laws or guidelines around employee data privacy makes it tricky for companies to know what to do. “It would be beneficial to have sort of those standards so that people could understand what you should and shouldn’t do. Today, especially in the U.S. and a lot of places in Asia, there’s sort of no laws around it,” said Waber.




We are still a long way from, “Alexa, do my taxes.”
Smart speakers are great for answering basic questions, like "what's the weather tomorrow" or "what's the population of Brazil" (it's 209.3 million, in case you were wondering). However, since the answers to many of these questions come from Google's ability to scrape the content of webpages, they can sometimes be inaccurate or purposely manipulated. Answers to questions about the UK government should now be fairly accurate, as the country has added "more than 12,000 pieces of information" to both Google Assistant and Alexa.




Old technology and the law.
EU lawmakers back wifi-based car standard in win for Volkswagen
The European Commission’s push for a wifi-based standard for cars backed by Volkswagen took a big step forward on Wednesday after EU lawmakers endorsed wifi over 5G technology promoted by BMW and Qualcomm.
… Wifi technology supporters include Renault, Toyota, NXP, Autotalks and Kapsch TrafficCom. The technology primarily connects cars to other cars.
… Fifth generation, or 5G, standard hooks up to both cars and devices in the surrounding environment, with a wider range of applications in areas such as entertainment, traffic data and general navigation.
The Commission has defended its stance on wifi technology, saying that it is available unlike 5G and that it would help to boost road safety.
Critics have said a requirement that new technologies be modified to be compatible with older technology is unrealistic and would put a brake on innovation.




There’s a joke here somewhere, but also (maybe) some truth.
Ukraine’s Comedian President Doesn’t Plan to Rule
The landslide victory of comedian and TV producer Volodymyr Zelenskiy in Sunday’s run-off presidential election in Ukraine poses a problem both for the country’s Western backers and those in the Kremlin who hope to exert control again. The Ukrainian people haven’t voted for a specific path, or even simply against politics as usual: They voted against being told what to do.
… Zelenskiy is Jewish, and he speaks better Russian than Ukrainian, which means most voters have proved unsusceptible to hardcore Ukrainian nationalism, which is traditionally anti-Semitic and intolerant to all things Russian. Zelenskiy is only 41, and he’s not a member of the post-Soviet political elite, a notoriously corrupt and self-serving bunch, and he’s a self-made millionaire with a successful production company.
But none of this is as important as why, and how, he won.
Poroshenko’s strategy had been to stress his dedication to nation-building: An alliance with the West, strengthening the military, gaining spiritual independence from the Russian Orthodox Church, supporting the Ukrainian language.



Sunday, April 21, 2019


Contrast with the FBI’s push to eliminate encryption, or at least provide them with a backdoor. Perhaps this kind of tool should be used by US campaigns to communicate.
Security flaw in French government messaging app exposed confidential conversations
The French government just launched its own messaging app called Tchap in order to protect conversations from hackers, private companies and foreign entities. But Elliot Alderson, also known as Baptiste Robert, immediately found a security flaw. He was able to create an account even though the service is supposed to be restricted to government officials.
Tchap wasn’t built from scratch. The DINSIC, France’s government agency in charge of all things digital, forked an open-source project called Riot, which is based on an open-source protocol called  Matrix.
In a few words, Matrix is a messaging protocol that features end-to-end encryption.
… Developing Tchap became essential as Emmanuel Macron’s campaign team relied heavily on Telegram — the French government still uses Telegram and WhatsApp for many sensitive conversations. By default, Telegram doesn’t use end-to-end encryption. In other words, people working for Telegram could easily read Macron’s conversations.




A most interesting topic. If you are going to die soon (according to the AI) will the AI Nurse just give you morphine? Lots and lots of morphine?
AI COULD PREDICT DEATH. BUT WHAT IF THE ALGORITHM IS BIASED?
EARLIER THIS MONTH the University of Nottingham published a study in PloSOne about a new artificial intelligence model that uses machine learning to predict the risk of premature death, using banked health data (on age and lifestyle factors) from Brits aged 40 to 69. This study comes months after a joint study between UC San Francisco, Stanford, and Google, which reported results of machine-learning-based data mining of electronic health records to assess the likelihood that a patient would die in hospital. One goal of both studies was to assess how this information might help clinicians decide which patients might most benefit from intervention.
The FDA is also looking at how AI will be used in health care and posted a call earlier this month for a regulatory framework for AI in medical care.




Clearly we will need some new names. “e-Antitrust” seems a bit unwieldy.
The Antitrust Case Against Facebook: a turning point in the debate over Big Tech and monopoly
In 2017, a 28-year-old law student named Lina Kahn turned the antitrust world on its ear with her Yale Law Review paper, Amazon's Antitrust Paradox, which showed how Ronald Reagan's antitrust policies, inspired by ideological extremists at the University of Chicago's economics department, had created a space for abusive monopolists who could crush innovation, workers' rights, and competition without ever falling afoul of orthodox antitrust law.
Now, Dina Srinivasan, a self-described technology entrepreneur and advertising executive who trained Yale Law School has done it again, with a magesterial, deftly argued paper for the Berkeley Business Law Journal called The Antitrust Case Against Facebook. It's one of the most invigorating, significant contributions to a new theory of antitrust for the digital age that I've ever read, ranking with Kahn's 2017 paper.
… Srinivasan shows how Facebook came to dominate our online discourse through activities that would have been prohibited under pre-Reagan theories of antitrust, and how, prior to these monopolistic tactics, Facebook was not able to conduct surveillance on its users, having to contend with multiple, bruising PR disasters and user revolts when it tried to do so.
[Interesting sub-title:
A MONOPOLIST’S JOURNEY TOWARDS PERVASIVE SURVEILLANCE IN SPITE OF CONSUMERS’ PREFERENCE FOR PRIVACY




Perspective. Perhaps this is why they are the next country to have a GDPR-like law…
Digital Life: Brazilians are Among the Most Avid Internet Users
This is the first series of infographics on Digital Life in Brazil.
  • More than two out of three Brazilians have access to smartphones and the internet
  • Brazilians spend more than 9 hours per day connected (among the highest rates in the world)
  • They rank very high in the world in social media use, including the use of Facebook, Instagram, YouTube, Netflix, WhatsApp, and Pinterest
  • Digital advertising continues to grow in double digits
  • E-Commerce, the sharing economy, and home delivery services are booming




Perspective. I would like to see cities owning the monopoly with anyone able to buy access.
Big Telecom Is Killing Your Rights To Municipal Internet Service In 26 States
High-speed internet and robust infrastructure is often the life blood of growing communities looking to attract high tech business and satisfy citizens. In most cases, however, residents and businesses only have access to one, or at the most two, broadband internet providers. You'll usually have access to broadband cable as one option and DSL as the other. So, municipal internet – where cities and towns build out and run their own broadband service – is an obvious solution.
BroadbandNow, which is a company that keeps tabs on broadband availability across the United States, has published a rather sobering study on the state of municipal broadband. Sadly, the company has discovered that over half of U.S. states (26 to be exact) have either outright banned the formation of municipal broadband networks or limit their expansion. While municipal internet seems like a win-win for Americans, big telecom is doing everything in its power to crush them at every turn.



Saturday, April 20, 2019


I went to the Privacy Foundation seminar yesterday. It helped me reach a few conclusions. Call them “Privacy Minimums” if you will:
First, Privacy Regulation will become universal. Today you need to consider the GDPR if you do business in the EU or with EU citizens. California and Brazil are mere months away from implementing their own, similar laws. Within the lifetime of a normal IT system, most countries will have similar regulations.
Second, none of the follow-on laws will make compliance simpler. You must comply with the toughest law, so why even consider “separate but simpler?”
Third, for each system (old or new) and for each data element, you will need to know where the data comes from, how it is used, everywhere it is stored and where (and when) it goes. Can it be used to identify an individual by itself or combined with other data? Do you have “opt in” permission? Has that individual ever exercised “opt out?”




How to handle a ransomware attack: back up in an hour!
A ransomware attack took The Weather Channel off the air
The Weather Channel was hit by a ransomware attack on Thursday, briefly taking a live TV program off the air, according to .a Wall Street Journal report
The attack came amid severe weather in the southeastern United States and knocked out the cable channel for more than an hour. The FBI told the Journal a ransomware attack was the source of the problem and that the agency is investigating.
“We experienced issues with this morning’s live broadcast following a malicious software attack on the network,” the channel confirmed in a tweet about the incident, adding that “backup mechanisms” had allowed the channel to restore service.




Three freedoms of the Internet?
Don't Regulate The Internet Like Every Company Is The Same
… key to his approach is a more modern update to the common "free as in speech v. free as in beer" concept that everyone in the open source world is familiar with. Ben talks about a third option that has been discussed for decades, which is "free as in puppy" -- meaning something that you get for free, but which then has an ongoing cost in terms of maintaining the free thing you got.



Friday, April 19, 2019


Hiding in the weeds.
Mueller report sheds new light on how the Russians hacked the DNC and the Clinton campaign
The Mueller report contains new information about how the Russian government hacked documents and emails from Hillary Clinton’s presidential campaign and theDemocratic National Committee.
At one point, the Russians used servers located in the U.S. to carry out the massive data exfiltration effort, the report confirms.
… The operatives working for the Russian intelligence directorate, the GRU, sent dozens of targeted spearphishing emails in just five days to the work and personal accounts of Clinton Campaign employees and volunteers, as a way to break into the campaign’s computer systems.
… By stealing the login details of a system administrator who had “unrestricted access” to the network, the hackers broke into 29 computers in the ensuing weeks, and more than 30 computers on the DNC.
… In all, some 70 gigabytes of data were exfiltrated from Clinton’s campaign servers and some 300 gigabytes of data were obtained from the DNC’s network.
… “I hope you’re able to find the 30,000 emails that are missing,” said then-candidate Trump at a press conference, referring to emails Clinton stored on a personal email server while she headed the State Department. Mueller’s report said “within approximately five hours” of those remarks, GRU officers began targeting for the first time Clinton’s personal office.




Big, but not a record.
Remember what I said earlier today about India being a data protection mess? Here’s another example. Mohit Kumar reports:
An unprotected database belonging to JustDial, India’s largest local search service, is leaking personally identifiable information of its every customer in real-time who accessed the service via its website, mobile app, or even by calling on its fancy “88888 88888” customer care number, The Hacker News has learned and independently verified.
Founded over two decades ago, JustDial (JD) is the oldest and leading local search engine in India that allows users to find relevant nearby providers and vendors of various products and services quickly while helping businesses listed in JD to market their offerings.
Rajshekhar Rajaharia, an independent security researcher, yesterday contacted The Hacker News and shared details of how an unprotected, publicly accessible API endpoint of JustDial’s database can be accessed by anyone to view profile information of over 100 million users associated with their mobile numbers.
Read more on The Hacker News,




Completely foreseeable. Outages will identify organizations that didn’t monitor the growth of the routing table.
Some internet outages predicted for the coming month as '768k Day' approaches
An internet milestone known as "768k Day" is getting closer and some network administrators are shaking in their boots fearing downtime caused by outdated network equipment.
The fear is justified, and many companies have taken precautions to update old routers, but some cascading failures are still predicted.
The term 768k Day comes from the original mother of all internet outages known as 512k Day.
512k Day happened on August 12, 2014, when hundreds of ISPs from all over the world went down, causing billions of dollars in damages due to lost trade and fees, from a lack of internet connectivity or packet loss.
The original 512k Day took place because routers ran out of memory for storing the global BGP routing table, a file that holds the IPv4 addresses of all known internet-connected networks.
… Many legacy routers received emergency firmware patches that allowed network admins to set a higher threshold for the size of the memory allocated to handle the global BGP routing table.
Most network administrators followed documentation provided at the time and set the new upper limit at 768,000 – aka 768k.




Another swing of the pendulum.
Nathan Sheard and Jennifer Lynch of EFF write:
Thanks to a recent ruling by Fairfax County Circuit Court Judge Robert J. Smith, drivers in Fairfax County, Virginia need not worry that local police are maintaining ALPR records of their travels for work, prayer, protest or play.
Earlier this month, Judge Smith ordered an injunction against the use of the license plate database, finding that the “passive” use of Fairfax County Police Department’s Automated License Plate Reader (ALPR) system violated Virginia’s Government Data Collection and Dissemination Practices Act (Data Act).
Read more on EFF.




People are responsible for Privacy Policies? What a concept!
Federal investigation of Facebook could hold Mark Zuckerberg accountable on privacy, sources say
Federal regulators investigating Facebook for mishandling its users’ personal information have set their sights on the company’s chief executive, Mark Zuckerberg, exploring his past statements on privacy and weighing whether to seek new, heightened oversight of his leadership.
The discussions about how to hold Zuckerberg accountable for Facebook’s data lapses have come in the context of wide-ranging talks between the Federal Trade Commission and Facebook that could settle the government’s more than year-old probe, according to two people familiar with the discussions. Both requested anonymity because the FTC’s inquiry is confidential under law.
… Often, the FTC does not target executives in cases where it finds a company’s business practices have violated web users’ privacy. But critics said that targeting Zuckerberg could send a message to other tech giants that the agency is willing to hold top executives directly accountable for their firms’ repeated data misdeeds.
“The days of pretending this is an innocent platform are over, and citing Mark in a large scale enforcement action would drive that home in spades,” said Roger McNamee, an early investor in the company and one of Zuckerberg's foremost critics.


(Related) How to bury bad news.
Facebook perfects the art of the news dump
On the Thursday before a major holiday weekend, and an hour before the much-anticipated Mueller report was released to the public, Facebook updated a month-old blog post titled "Keeping Passwords Secure" with a few lines of italicized text.
"Since this post was published, we discovered additional logs of Instagram passwords being stored in a readable format. We now estimate that this issue impacted millions of Instagram users," says the update.
The original post revealed Facebook stored passwords for hundreds of millions of its Facebook users and "tens of thousands" of Instagram users as plain text in a database that could be accessed by its staff.




Almost three years now and some people still haven’t read it?
GDPR Article 27 … The ‘Unknown Obligation’ of Appointing a Nominated European Representative
… Whilst the GDPR is a European regulation, many organizations outside of Europe will be unaware that they are required to appoint a Nominated European Representative under certain conditions (as per Article 27 of the GDPR).


(Related) Dead because tech companies had input?
Hunton Andrews Kurth writes:
The much-discussed Washington Privacy Act, Senate Bill 5376 (“SB 5376”), appears to have died after failing to receive a House vote by an April 17, 2019 deadline for action on non-budget policy bills. Though the bill could be revived before the regular session ends on April 28, 2019, Washington lawmakers expressed doubt.




I’m detecting a strong anti-AI bias…
Some AI just shouldn’t exist
Human bias can seep into AI systems. Amazon abandoned a recruiting algorithm after it was shown to favor men’s resumes over women’s; researchers concluded an algorithm used in courtroom sentencing was more lenient to white people than to black people; a study found that mortgage algorithms discriminate against Latino and African American borrowers.
The tech industry knows this, and some companies, like IBM, are releasing “debiasing toolkits” to tackle the problem. These offer ways to scan for bias in AI systems — say, by examining the data they’re trained on — and adjust them so that they’re fairer. [A great entry point for hackers. Bob]
But that technical debiasing is not enough, and can potentially result in even more harm, according to a new report from the AI Now Institute.
The three authors say we need to pay attention to how the AI systems are used in the real world even after they’ve been technically debiased. And we need to accept that some AI systems should not be designed at all.
… In other words, ensuring that an AI system works just as well on everyone does not mean it works just as well for everyone.




Attempts to have AI interpret politicians caused the AI to stroke out.
A neural network can read scientific papers and render a plain-English summary
… a form of artificial intelligence (AI) ... can read scientific papers and render a plain-English summary in a sentence or two.
Even in this limited form, such a neural network could be useful for helping editors, writers, and scientists scan a large number of papers to get a preliminary sense of what they're about.




Interesting. Managers don’t want to listen to their lawyers?
The Rise of Risk Management in Financial Institutions – Diminution of Legal Function
Business Law Today – The Rise of Risk Management in Financial Institutions and a Potential Unintended Consequence – The Diminution of the Legal Function By: Thomas C. Baxter, Jr. After the global financial crisis, a highly respected group of financial supervisors from the industrialized world convened to consider what might have caused the worst financial crisis experienced since the Great Depression. This group – aptly named the “Senior Supervisors Group” – concluded that a material contributing cause was what they characterized as a “colossal failure of risk management.” The Senior Supervisors Group was not alone. Many other bodies have taken up the same topic and reached a similar conclusion. In the 10 years since the global financial crisis ended, the financial community has responded to the identified causes of the financial crisis, adopting lessons learned and significantly reforming the financial system. This work has resulted in a financial system with individual institutions that are demonstrably more safe and more sound than before, and a much more resilient banking system overall. In contrast to what existed on the eve of the crisis – early 2007 – today’s financial system has considerably higher capital and liquidity, as government officials and other commentators have observed. In addition, and perhaps even more importantly if we accept the conclusion of the Senior Supervisors Group, there has been a revolution in the discipline of risk management and in the “build-out” of processes and procedures for identifying, measuring, monitoring, and controlling risk. In the United States, for example, one may witness the Dodd-Frank Wall Street Reform and Consumer Protection Act, which President Obama signed into law on July 21, 2010 (the “Dodd-Frank Act”). The Dodd-Frank Act introduced varied and different requirements for risk management, including a series of “enhanced prudential standards,” as well as governance directed at risk management requirements, like the requirement for a risk committee of the board of directors….
This article will discuss whether the rise of the risk management function has had one very specific unintended consequence – the diminution of the legal function. To place such an important question in a proper context, this article will focus on the potential inverse relationship – it is not only that the legal function has declined in importance, but it is also that the decline has come as the direct result of the rise in risk.