P1: Protecting Privacy: Consent and Anonymization

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/8

encourage image

There's no tags or description

Looks like no tags are added yet.

9 Terms

1
New cards

What is informational privacy? What are the two views about why informational privacy matters that we considered (J. Rachels and M. Marlinspike)?

Informational Privacy: The ability to control who has access to personal data.

J. Rachels: privacy enables us to control our social relationships.

M. Marlinspike: privacy is a necessary defense against government oppression.

2
New cards

What are three strategies for protecting the privacy of personal data (consent, anonymization, and differential privacy)? How is each strategy defined?

Consent: requires getting explicit user agreement after explaining data use.

Anonymization: strips personal identifiers from datasets.

Differential Privacy: adds statistical noise to individual records to protect privacy while preserving overall population data.

3
New cards

What is the first problem with the consent approach (no one reads terms and conditions)? Why does this problem suggest that agreeing to terms doesn't mean maintaining control?

Privacy policies are far too long and complex for people to read before agreeing. Because users sign unread terms, agreeing to them fails to grant users actual control over their personal data.

4
New cards

What is the transparency paradox? What is the two-step argument that it shows we cannot give informed consent?

The transparency paradox states that data practices cannot be explained both accurately and simply enough for average users to comprehend. Since companies cannot explain practices simply and accurately, users cannot provide genuine informed consent.

5
New cards

What is de-identification (anonymization), and what is its goal? Why is it supposed to remove privacy concerns?

De-identification removes personal links from data so individual identities cannot be determined. It is thought to eliminate privacy concerns because unlinkable data prevents anyone from inspecting your private life.

6
New cards

What is personally identifiable information (PII)? What are some examples?

PII is any data that can directly or indirectly identify an individual. Examples include legal names, Social Security numbers, physical home addresses, and email addresses.

7
New cards

What is the difference between full anonymization and pseudonymization? What is an example of pseudonymization?

Full anonymization permanently strips all identifying details, whereas pseudonymization replaces direct PII with a unique tracking key (e.g., Google's Advertising ID).

8
New cards

What are the two problems for the pseudonymization approach discussed in lecture?

The two main problems are user "reachability" and vulnerability to reidentification attacks.

9
New cards

What does it mean to be "reachable" with respect to information about you? Why doesn't pseudonymization prevent companies from "reaching" us?

You are reachable if institutions can use your data to target, profile, or sanction you. Pseudonymization fails to stop reachability because persistent tracking keys allow platforms to profile behavior and deliver targeted actions like ad exclusions.