Automated Decision Making (OCR A-Level Computer Science): Flashcards
📚Flashcards
Practise the cards
14 cards from this deck
ShowHide
Practise the cards
14 cards from this deck
Automated decision-making definition
Automated decision-making definition
Computer systems making decisions without human intervention
Four main areas using automated decision-making
Four main areas using automated decision-making
Finance, healthcare, recruitment, law enforcement
Moral issue: Accountability question in automated systems
Moral issue: Accountability question in automated systems
Who is responsible: developer, organisation, or machine?
Social issue: Impact of automation on employment
Social issue: Impact of automation on employment
Automated systems replace human roles, affecting employment
Social issue: Digital literacy disadvantage
Social issue: Digital literacy disadvantage
People without digital literacy or resources may be disadvantaged by systems
Ethical issue: Cause of bias in algorithms
Ethical issue: Cause of bias in algorithms
Training on biased data (e.g., racial/gender bias in hiring)
Ethical issue: Informed consent requirement
Ethical issue: Informed consent requirement
People should be informed when automated systems use their personal data
Cultural issue: Cultural sensitivity in AI systems
Cultural issue: Cultural sensitivity in AI systems
Systems must account for cultural differences and values
Opportunity: Efficiency benefit of automated systems
Opportunity: Efficiency benefit of automated systems
Process large amounts of data quickly for faster decisions
Opportunity: Consistency advantage
Opportunity: Consistency advantage
Reduces human error and bias, offering consistent outcomes
Risk: How bias is replicated in automated systems
Risk: How bias is replicated in automated systems
Training data biases are replicated and potentially amplified
Risk: Privacy concern with automated decision-making
Risk: Privacy concern with automated decision-making
Extensive data collection raises misuse and privacy concerns
Risk: Loss of human oversight consequence
Risk: Loss of human oversight consequence
Lack of nuanced judgement and empathy, leading to unfair outcomes
Common mistake: Assumption about algorithm neutrality
Common mistake: Assumption about algorithm neutrality
Assuming algorithms are neutral; they reflect data and design