Emerging Issues

In practice, 2 themes are especially important in applying privacy protections to big data...

1- data minimization, to avoid privacy and security risks of big data where possible and 2-deidentification, to avoid privacy and security risks that arise when previously deidentified data can be reidentified.

IoT concerns regarding privacy and cybersecurity

Concerns regarding privacy and cybersecurity stem from (1) limited user interfaces in the products, (2) lack of industry experience with privacy and cybersecurity (3) lack of incentives in the industries to deploy updates after products are purchased, and (4) limitations of the devices themselves, such as lack of effective hardware security measures.

DDoS

Disruptive Denial of Service attack - A DoS or DDoS attack is analogous to a group of people crowding the entry door of a shop, making it hard for legitimate customers to enter, thus disrupting trade. Criminal perpetrators of DoS attacks often target sites or services hosted on high-profile web servers such as banks or credit card payment gateways.

Mark Rotenberg (Electronic Privacy Information Center (EPI))

In a synthesis of Asimov's principles with modern concepts of privacy, Mark suggested 2 additions to Asimov's iconic laws to address transparency in algorithms and AI. These are as follows: 1 - Robots should always reveal the basis of their decision. 2- Robots must always reveal their identities

Moore's Law

Intel Co-Founder Gordon Moore predicted the exponential increase in computing power needed to handle the massive amounts of data, and science fiction writer Isaac Asimov discussed the social and ethical implications of the cognitive computing power that would come from such rich data analysis. In 1965, Moore published a now iconic article in which he observed that the number of transistors that would fit onto a circuit board doubled each year. A decade after this publication, the principle, dubbed "Moore's Law" was tweaked to say that the number of transistors on a circuit board doubled every 18 to 24 months. For most non-engineers, this law is understood to say that computing power doubles every 18 months. Moore's Law is not a mathematical law; it's a useful simplification about the exponential growth in computing for the last several decades. This law is a useful way to explain the development of two of the technological phenomena emerging at the time of the writing of this book - big data and IoT.

IoT

Internet of Things - represents a new development in the ways that individuals interact with computing devices. Things like smart homes, watches, connected cars, wearable technology, etc. There is not one universally accepted definition of internet of things. A critical aspect of IoT is that the devices can connect to the internet (and each other) without the need for human interaction. The FTC has examined privacy and security issues using the following definition of IoT; Devices or sensors - other than computers, smartphones, or tablets - that connect, communicate, or transmit information with or between each other through the internet. Most IoT devices share two characteristics that are important for privacy and cybersecurity discussions: (1) the devices interact with software running elsewhere (often in the cloud) and function autonomously and (2) when coupled with data analysis, the devices may take proactive steps and make decisions about or suggest next steps for users.

Emerging Field of Internet of Robotic Things (IoRT)

IoT and robotics have traditionally been considered separate fields. A new field though called Internet of Robotic Things (IoRT) is emerging. In the most basic terms, the machine-to-machine communication of gathered data in Iot is combined with the physical aspects of robotics. This field brings with it many of the privacy and security concerns of both IoT and robotics. for IoRT, categories of privacy risks include interactions with individuals and use of sensors. Security concerns include communications between user and robot, authentication of users, lack of encryption of data and ease of accessing programmable features.

Data minimization

More is not better! Lots of data can lead to a huge data breach. Companies must also provide access control - making sure (making sure only employees with a need to know can access the personal information) and segregation of databases, to prevent one lapse from harming the entire company.

National Highway Traffic System Administration (NHTSA)

Their mission is to save lives, prevent injuries, and reduce economic costs due to road traffic crashes, through education, research, safety standards, and enforcement.

FTC Report: Data Brokers: A Call for Transparency and Accountability

2014 - The FTC expressed concern about the vast amount of personal information collected by data brokers. These data brokers collect numerous sources of data usually without consumers' consent. They analyze data about consumers to draw inferences about them; and combining online and offline data to market to consumers online. The FTC identified 3 broad categories of products offered by data brokers at that time: (1) Marketing (appending data to customer information that a marketing company already has) (2) risk mitigation (such as information that may reduce the risk of fraud) (3) location of individuals (such as identification of an individual from partial information). For each of these instances, the FTC suggests that data brokers engage in data minimization practices, review collection practices carefully as they relate to children and teens, and take reasonable precautions to ensure that downstream users do not use the data for discriminatory or criminal purposes.

FTC Report on IoT

2014, the FTC undertook its first enforcement action involving an internet-connected device against a company that provided consumers with internet-connected cameras for use inside the home. In its complaint, the FTC alleged that TRENDnet failed to encrypt customer log-in credentials and failed to test consumers' privacy settings. Hackers utilized these security vulnerabilities to post hundreds of live video feeds featuring babies sleeping in cribs and adults engaging in daily activities. The agency's complaint resulted in an order requiring the company to bring its practices into line with FTC requirements and to establish its compliance by undergoing assessments every two years for the next 20 years. One year after this - the FTC issued the report titled "Internet of things: Privacy and Security in a Connected World"...

FTC Report: Big Data: A Tool for Inclusion or Exclusion?

2016 - This report discussed the era of big data as being brought about by dramatic reduction in the cost of storage, collection of consumer data from almost innumerable sources (as discussed in the IoT section), and increase in computer capabilities to analyze this data. It noted that this era of big data is still in its infancy but that analytics resulting from big data already benefited areas such as marketing, human resources, and fraud prevention. This depicted is benefits and risks. Benefits include providing healthcare tailored to individual patients, enhancing educational opportunities by tailoring the experience to the individual student, and increasing equal access to employment. Risks include exposing sensitive information, reinforcing existing disparities, and creating new justifications for exclusion. Issues of concern where privacy of sensitive information, security of large sets of data, and disparate impacts on individuals from inappropriate or inaccurate inferences drawn from analytics. The FTC cautioned companies that numerous federal laws already applied to handling big data: the Fair Credit Reporting Act (FCRA), the Equal Credit Opportunity Act (ECOA) and the Federal Trade Commission (FTC) Act.

IoT Connected Cars

Another IoT device that this book examined. These collect and transmit data about the vehicle, the driver's driving habits, and the driver's preferences. One example would be a vehicle that wirelessly alerts the dealership when tires need to be rotated. Another would be an app from a car insurance company that records braking habits. Additionally, information may be transmitted from the car to the internet from multiple sources, such as users' phones, video systems, cameras, GPS systems, and entertainment centers. This will continue to increase as cars become even more autonomous. These various systems have different levels of interoperability and security and privacy experts raise concerns that these configurations place sensitive information at risk to unauthorized access or hacking. The complexity of these issues has raised federal agencies to consider regulating connected cars; the National Highway Traffic System Administration (NHTSA), the FTC, and the Federal Communications Commission (FCC). In 2017, the FTC and the NHTSA and the US Department of Transportation (DOT) issued new federal guidance for automated vehicles. Although this document did not directly mention privacy, the related material noted that the FTC has the responsibility to protect consumer privacy and that the 2 federal agencies would continue to work together on matters involving motor vehicle safety and consumer privacy. In 2018, the FTC released a document that included the key takeaways from the Connected Cars Workshop in regards to this technology and what is collected, security risks, etc.

IoT Smart Homes

Another IoT device that was examined was Smart Homes. Smart Homes typically have multiple devices connected to the internet to enhance the home environment experience. These can include refrigerators, thermostats, beds that adjust, etc. This raises a lot of Privacy concerns obviously. These issues are intensified even further by the massive amounts of data collected, the fact that much of the data is reported back to companies over the internet (often without the awareness of the user), and the reality that these systems can be hacked or hijacked (and often the data streams are not encrypted). In 2019, Apple, Amazon and Google announced a collaboration called Project Connected Home over IP. One of the goals of the project is to develop an open-source connectivity standard. Despite benefits to consumers, who would experience a more seamless experience between appliances and smart assistants, privacy experts warn that companies will have access to more data, which would allow increasingly detailed profiles and additional surveillance.

Broadband Internet Technical Advisory Group Report

BITAG - In 2016, the BITAG issued a set of recommendations for IoT privacy and security practices. The main recommendations include: - IoT devices should follow security and encryption best practices - For devices that can be customized by the users, the company should test the IoT devices in different possible configurations. - IoT devices should be designed to facilitate automated, secure software updates - IoT devices should be shipped originally with reasonably up-to-date software - IoT device should be shipped originally with a privacy policy that is understandable and easy to find - IoT devices should communicate with restrictive rather than permissive protocols - IoT devices should continue to function if internet connectivity is disrupted or if cloud backup fails.

Data deidentification

Big data can call into question the traditional line between personally identifiable information, which is subject to a more stricter privacy rule, and deidentified information, where privacy rules traditionally no longer apply. A goal for organizations should be to gain the benefits from analyzing data while seeking to suppress the portions of records that can reveal individuals' identities.

Asimov's Law of Robotics

Big data is the fuel that runs algorithms and analytics, which will enable artificial intelligence (AI) systems connected to the cloud. In the 1940's, Asimov spoke in terms of robots. In today's world, these emblematic laws apply not just ot robots but to algorithms, machine learning, and AI (meaning that Asimov's three "laws", to the extend they apply in practice, will apply to the practices in IoT, big data and the cloud. These 3 laws are: 1 - A robot may not injure a human being or, through inaction, allow a human being to come to harm. 2 - A robot must obey orders given it by human beings except where such orders would conflict with the First law. 3 - A robot must protect its own existence as long as such protection does not conflict with the First or Second Law. Later, Mark Rotenberg suggested 2 additions to these iconic laws (depicted further below).

Satya Nadella (Microsoft CEO)

Paying homage to the starting point provided by Asimov, Microsoft's CEO Satya Nadella proposed a list of AI design principles, including several that focus on privacy issues: - AI must be designed to help humanity - AI must be designed for intelligent privacy - AI must be transparent, and - AI needs algorithmic accountability so humans can undo unintended harm.

FTC has issued reports that examine issues that arise as a result of the use of big data.

Privacy practitioners are encouraged to become familiar with the key points in these reports, as they can indicate areas where the federal agency will focus its future efforts. These reports are detailed in upcoming slides.

IoT Smart Cities

Smart Cities is a term that primarily refers to municipalities and other government entities using sensors to monitor functions and improve government services. For example, a city could embed wirelss sensors into existing lighting fixtures, then analyze this data to allow targeted law enforcement practives, improved parking efficiency, and increased environmental monitoring. This could help manage everything from traffic to garbage collection and parking meters, etc. And is expected to grown . This can raise concerns about software vulnerabilities, data security breaches, and potential invasions of privacy. Also, since this is government and not personal - it raises even further concerns. In 2015, the US Department of Homeland Security (DHS) issued a report highlighting 3 themes in cybersecurity risk that arise when integrating cyber-physical systems with city infrastructure in smart cities.

FIPPs

The Fair Information Practice Principles (FIPPs) are a set of internationally recognized principles that inform information privacy policies both within government and the private sector. used by numerous foreign countries and international organizations. EU - FIPPS are accounted for within GDPR US - FIPPS are often incorporated within a companies privacy policies, and violation of those policies can lead to enforcement under the FTC. Relevant FIPPs include: - Collection Limitation - obtain data by lawful and fair means and (where possible) with the knowledge and consent of the data subject. - Purpose specification - use the data as initially intended or for other purposes that are not incompatible with that purpose (meaning there are limitations on secondary uses of the data) - Use limitation - do not use or disclose data beyond those purposes except with consent or by authority of law.

IoT Wearables

This book examines 3 IoT devices. Wearables are electronic devices that are worn on the body and collect data in real time. These can range from headwear used by soldiers on the battlefield to wrist devices that check heartbeats during exercise. Analysis of the data from these devices presents the possibility of benefits as well as significant challenges. Benefits are easily seen but privacy challenges exist when the devices collect health-related information such as heart-rate, blood pressure, and even more complex information such as blood alcohol content. Most of this information is NOT protected by HIPAA, because HIPAA applies only to the activities of covered entities such as providers and health insurance plans. These challenges related to wearables data have been examined in research that focused on users' privacy concerns regarding wearables and things like "Right to forget", Impact of location disclosure, Concern that screens will be read, Video and audio recording of unknowing subjects and others. In 2016, the Future of Privacy Forum issued a set of best practices for the privacy of consumer wearables. These include access, deletion, and correction rights; opt-in consent for sharing with 3rd parties; sharing of data for scientific research purposes, with informed consent; compliance with leading app platform standards and global privacy frameworks; strong data security requirements; and strong requirements for deidentification. In 2017, the FTC provided advice to consumers on how to secure their smart watches by setting a PIN, selecting a lock pattern, and locking the watch if it is too far from the associated phone. In 2018, the FTC issued warning letters to 2 smart watch companies that target their products to children - explaining that the companies did not appear to be providing appropriate notice or obtaining parental consent as required by COPPA.

Big Data

This is a term used to describe the nearly ubiquitous collection of data about individuals form multitudinous sources, coupled with the low costs to store such data and the new data mining techniques used to draw connections and make predictions based on this collected information. On a positive side, big data provides the basis for modern analytics and the significant insights that can be derived from such data - often by means of machine learning. On the cautionary side, big data and modern analytics can be a difficult fit for fair information privacy practices (FIPPs) sometimes called fair information practice (FIPs) because there may not be clear notice to how data is used, and advanced analytics may not be written within the purposes the individual expected when the data was collected. Underpinnings of the broad term for big data are things like analytics programs, algorithms, machine learning, and other data mining techniques. This information is gathered by different devices know as IoT. Big Data is categorized by 3 V's: - Velocity (how fast the data is coming in) - Volume (the amount of data coming in) - Variety (what different forms of data are being analyzed)

FTC report, "Internet of things: Privacy and Security in a Connected World"

This is for the home based technology with regard to privacy of personal information, the FTC pointed out that the sheer volume of information collected as well as the deeply personal nature of the information obtained heightened the need for protection. The FTC explained that traditional models of providing consumer disclosure and choice may need to be modified by companies involved in this industry. The FTC is keenly aware of the practical difficulty of providing choice when there is no consumer interface and suggests that companies look at utilizing one or more of the following: choice at point of sale, video tutorials, codes on the devices, choice during setup, management portals or dashboard, etc. The also detailed security risks that allow intruders access to personal information collected by the devices.

Masking

This technique masks the original values in a data set with the goal of data privacy protection. One way this may be accomplished is to use perturbation - make small changes to the data while maintaining overall averages - to make it more difficult to identify individuals.

Blurring

This technique reduces the precision of disclosed data to reduce the certainty of individual identification. For example, date of birth is highly identifying (because a small portion of people are born on a particular day of a particular year), but year of birth is less identifying. Similarly, a broader set of years (such as 1971-1980 or 1981-1990) is less identifying than year of birth.

Differential Privacy

This technique uses a mathematical approach to ensure that the risk to an individual's privacy is not substantially increased as a result of being part of the database.

Big Data Case Study

To understand the concerns that arise from "Big Data" the book discusses a case study of a major financial services firm that decides to create an "All Customer Funds" (ACF) database, showing all transactions and balances for customers' checking accounts, savings accounts, securities, real estate, and other assets. The ACF software collects a tremendous amount of data, with granular financial information about tens of millions of customers. Data at this scale enables advanced analytics, with potentially great benefits to customers (higher ROI) and the bank (greater profit per customer). Along with benefits, comes with risks. The bank will not be pleased if the big data products turns into a big data breach - a centralized database can create the possibility of larger breaches than previously. A breach can be expensive, due to the costs of responding to the breach (such as notice to customers under data breach laws), fraudulent account activity (when a hacker is able to withdraw money from customer accounts) and identity theft. Information security is paramount. The scale of a larger database creates a target for hackers.