Week 13 - Surveillance and Censorship Flashcards
(31 cards)
Surveillance
Systematic observation or data collection concerning people with the aim of influencing or managing their behaviour.
Key concepts of surveillance
Consent: are we aware and okay with being watched?
Power: who has the authority to watch/ access data and is there a way to remove consent from being watched?
Data: what is being collected and how is it used?
Types of surveillance
- State/Government surveillance.
- Corporate surveillance.
- Personal surveillance.
- Self surveillance.
Covert vs overt surveillance
Covert surveillance: Techniques used discreetly so the subject is unaware of being monitored.
Overt Surveillance: Visible and recognizable monitoring methods
State/government surveillance uses
- national security
- law enforcement and crime prevention
- public safety (e.g., counterterrorism)
State/government surveillance issues
- Privacy violations: lots of data collected, often without informed consent.
- Power imbalance: government holds vast data, citizens have little oversight.
- Overreach and abuse: risk of targeting dissidents, indefinite data retention
Corporate surveillance uses
- Profit motive: selling behavioural data, optimizing adds.
- Consumer profiling: predicting preferences and tailoring marketing.
- Productivity oversight: monitoring employees for efficiency.
Corporate surveillance issues
- Lack of consent/ transparency: users rarely realise how much is tracked.
- Data monetisation: data may be sold to third parties.
- Ethical and legal concerns: biased analytics and manipulative recommendation systems.
Personal surveillance uses
- Safety: child protection, home security.
- Personal convenience: home deliveries, letting family know whereabouts.
- Peace of mind: tracking personal belongings.
Personal surveillance use issues
- Consent and boundaries: monitoring someone else (spouse/child) can erode trust.
- Misuse and abuse: stalkerware and controlling behaviour in domestic contexts.
- Data security: Personal devices are susceptible to hacking or data leaks.
Self surveillance uses
- Self improvement: health goals, productivity.
- Personal insight: tracking habits, measuring performance.
- Sharing achievements: gamification (elements of games being used in other contexts), social bragging rights.
Self surveillance issues
- Data privacy: personal health metrics stored on corporate servers.
- Over-monitoring: obsession with metrics may create anxiety or skew behaviour.
- Commercial exploitation: collected data can be resold or used for advertising.
USA Patriot Act (2001)
Signed into law after 9/11 terror attacks to allow searching of emails and telephone records without a warrant. Openly conducted surveillance on US and foreign citizens.
Criticised for not informing of consent and did not follow “innocent until proven guilty”.
Mass surveillance
The practice of spying on a significant part of the population.
E.g., GCHQ used Karma Police to access website metadata, stored it in a repository called Black Hole and used a tool called Mutant Broth which allowed searching of Black Hole. This violated the legal principle of probable cause.
UK surveillance legislation
- Anti-Terrorism, Crime and Security Act, 2001: enabled retention of communication data voluntarily but does not include content of communications.
- Communications Data Bill, 2012 (Snooper’s Charter): Requires all ISPs to store user data for 12 months “to catch criminals and protect children”.
- Investigatory Powers Bill, 2016 (Snooper’s Charter 2.0): Enables bulk collection of data and requires companies to assist in bypassing encryption.
Big data surveillance
The systematic collection, analysis, and use of massive datasets for monitoring and control.
Enables predictive policing, counterterrorism strategies, and broader control of populations through pattern recognition.
Application areas for big data surveillance
- National security: predicative models for identifying potential threats.
- Law enforcement: real-time data from internet of things (IoT), CCTV and AI-driven analytics.
- Corporate security: protecting assets and monitoring employees.
Processing techniques for big data surveillance
- Machine Learning (ML): algorithms for behavioural analysis.
- Natural Language Processing (NLP): used for monitoring communications.
- Graph theory: used to map social networks e.g., identifying key influencers in a network.
Predictive intelligence
- Big data analytics can anticipate events, such as potential crimes or terrorist attacks, using historical data.
- E.g., PredPol is a predictive policing tool that analyses crime patterns and deploys resources proactively.
- Can be used for network anomaly detection, fraud detection in financial systems and insider threat detection within organisations.
Sousveillance:
- The practice of individuals monitoring those in power, such as governments, corporations, or other authorities.
- E.g., recording police during protests or public events, whistleblowing to expose missuses of power (Edward Snowden).
- Sousveillance is intended to empower individuals to hold authorities accountable and challenge abuses of surveillance systems.
Nothing to hide argument for surveillance
It can be possible to find someone guilty of something even when they are innocent.
Possible through:
- Distortion: surveillance can make someone seem guilty through misinterpretation of data pr framing innocent behaviours as suspicious.
- Exclusion: surveillance systems often prevent people from knowing how their data is being used or correcting inaccuracies - errors can misrepresent individuals as criminals.
Traditional vs Digital censorship
- Traditional censorships: blocking books, banning films or controlling media broadcast.
- Digital censorship: automated systems filtering content, blocking websites or suppressing dissenting opinions online.
Actors of censorship
- State actors: governments imposing restrictions to control public discourse (e.g., China’s great firewall).
- Corporate actors: platforms like YouTube, Facebook etc. censor misinformation, hate speech and politically sensitive topics.
- Algorithmic moderators: AI systems tasked with removing harmful content. Can often result in unintended censorship due to biases.
Types of censorship
- Network level censorship: blocking websites or services through DNS tampering, IP blocking or deep packet inspection.
- Platform level censorship: content moderation on platforms like Twitter, Facebook, YouTube by using algorithms to detect and remove flagged content.
- Self-censorship: Individuals modify behaviour knowing they are being monitored or flagged (linked to surveillance).
- Algorithmic censorship: AI filters unintentionally remove content due to training bias or lack of contextual
understanding
