Topic 8: Big Data & Machine Learning: Ethical & Privacy Issues Flashcards

1
Q

Discuss the potential big data for business.

A

Potential:

  • Source of innovation
  • Enabler of development
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Explain how the new term “data trust deficit” developed.

A

public developed greater awareness and sensitivity towards the big data topic driven by increasingly prominent role of IoT.

Big Brother serves the economic interest of businesses over which people have little or no control.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

List five methods of protecting human rights in the ‘Era of Big Data.

A
  1. Stop High-Tech profiling
  2. Ensure fairness in Automated Decisions
  3. Respect the Law
  4. Enhance Individual Control of PI
  5. Protect People from Inaccurate Data
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Describe a concern for each of the three main privacy issues: customer profiling, group privacy, and data security.

A

Customer profiling: individuals have no way to contest the information that has been gathered about them.

Group privacy: could be used for ethically questionable marketing strategies (e.g. target racial group).

Data security: misuse of data by employees

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Discuss what constitutes informed consent.

A

Individuals agree to the terms & conditions BEFORE their data can be used for research of commercial purposes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

List six questions that ethics professionals within an organization using big data can ask themselves.

A
  1. Do we know how the company uses Big Data and to what extent it is integrated into strategic planning?
  2. Do we send a privacy notice when we collect personal data? Is it written in a clear and accessible language which allows users to give a truly informed consent?
  3. Does my organisation assess the risks linked to Big Data?
  4. Does my organisation have any safeguard mechanisms in place to mitigate these risks?
  5. Do we make sure that the tools to manage these risks are effective and measure outcome?
  6. Do we conduct appropriate due diligence when sharing or acquiring data from third parties
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

List three main features characterizing artificial intelligence

A
  1. Learning - ability to acquire relevant/rules information
  2. Reasoning - ability to turn Learning into outcomes
  3. Iterative - Ability to adapt based on new information
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

List three immediate risks of artificial intelligence

A
  1. Ethics risk - uses might lead to ethical lapses
  2. Workforce risk - automation lead to deskilled workers
  3. Technology risk - black box algos can make it difficult to identify tampering
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Discuss each of the following as they impact the ethical nature of applications of artificial intelligence in business: Accurate results

A

Companies need to ensure AI systems they use produce correct, precise, and reliable results.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Discuss each of the following as they impact the ethical nature of applications of artificial intelligence in business: Respect for privacy

A

AI is often coined as the death of privacy. European commission says that everyone has the right to protection of personal data, this led to GDPR.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Discuss each of the following as they impact the ethical nature of applications of artificial intelligence in business: Transparency and openness

A

Usually the code for algos are kept secret, being transparent is an important step. (improves trust)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

List five measures organizations can take to minimize the risk of ethical lapses due to improper use of AI technologies.

A
  1. Design new and more detailed decision-making tools (ensure they do the right thing and are in line with company’s ethical values)
  2. Engage with third parties (if ethical values are same)
  3. Establish Ethics Research Unit.
  4. Introduce ‘ethics tests’ for AI machines
  5. Empower people through training
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

List potential questions addressing the use of AI in a code of ethics.

A
  1. What is the purpose of our job and what AI do we need to achieve it.
  2. Do we understand how these systems work?
  3. Are we in control of this technology?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Describe the primary purpose of the GDPR.

A

seeks to harmonize the protection of fundamental rights and freedoms of natural persons in respect of processing activities and to ensure the free flow of personal data between Member States

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Describe the key changes in data protection regulation, including the meaning of
• Rights of the individual

A

Individuals have the right to access, amend, restrict, withdraw consent and request that their personal data be erased.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Describe the key changes in data protection regulation, including the meaning of
• Informed Consent

A

Requests for consent to be explicit (not implied) and written in clear and easy to understand language, and as easy to withdraw as to give.

17
Q

Describe the key changes in data protection regulation, including the meaning of
• Notification

A

Mandatory breach notification period of 72 hours.

18
Q

Describe the key changes in data protection regulation, including the meaning of
• Data portability

A

Customers can transfer personal data from one company to another.

19
Q

Describe the key changes in data protection regulation, including the meaning of
• Supervision and enforcement

A

One stop shop approach; higher penalties (max 20meur or 4% of revenues)

20
Q

Describe the key changes in data protection regulation, including the meaning of
• Liability

A

Data processors, in additional to the Data Controller, are directly liable

21
Q

Distinguish between two types of threats of personal data breaches

A

External and Internal

22
Q

Discuss ‘people risk.’

A

poorly trained individual with a desire or capacity to act against values.

23
Q

List key questions around the role an ethical culture plays in preventing data breaches.

A
  1. Have we decided at Board level and throughout the organization what constitutes an appropriate attitude and approach in choosing how to respond to threats and breaches?
  2. Is our culture sufficiently built on our values such that each individual knows and understands why and how we approach confidentiality and privacy the way we do?
  3. Do we make our decisions and state outcomes with openness, transparency and honesty?
24
Q

Describe how establishing the boundaries and standards can help foster an ethical culture and compliance with the GDPR.

A

It provides clarity on the company’s approach. What should people expect within the process and what can they do if they are not satisfied.

25
Q

Identify appropriate tools for implementing sound ethical practices.

A

Implement a check list

26
Q

Recognize the main point behind each of the five “C’s”: Consent

A

At every step of building a data product, it is essential to ask whether appropriate and necessary consent has been provided.

27
Q

Recognize the main point behind each of the five “C’s”: Clarity

A

Users cannot really consent to anything unless they know clearly what they are consenting to.

28
Q

Recognize the main point behind each of the five “C’s”: Consistency

A

Trust requires consistency over time.

29
Q

Recognize the main point behind each of the five “C’s”: Control & Transparency

A

Once you have given your data to a service, you must be able to understand what is happening to your data.

30
Q

Recognize the main point behind each of the five “C’s”: Consequences and harm

A

It is essential to ask whether the data that is being collected could cause harm to an individual or group (strava -> US military)

31
Q

Explain how to implement the five C’s.

A

It should be part of the organizations culture, the company should consider developing a checklist before releasing a product to the public.

32
Q

Identify major issues in ethics and security training.

A
  • in other fields ethics is an essential part of professional education not with data science
  • Ethics is often an elective and not embedded in other classes. (isolated)
  • Security is often an after thought
33
Q

Argue for a method of developing guiding principles

A

A checklist with a short set of questions, you don’t go to the next stage w/o answering the questions.

34
Q

Describe how to build ethics into a data-driven culture. Identify four methods for doing so.

A
  1. Empower the individual to stop the process before damage is done (Toyota’s assembly line)
  2. Anyone should be able to escalate issues w/o fear
  3. An ethical challenge should be part of the hiring process.
  4. Product reviews must ask questions about the product’s impact.
35
Q

Discuss the regulatory environment for data and new technologies in terms of consent.

A

Laws and regulations almost always lag behind technology. Also the committees that make policy often lack experts with the right technical background.