AIGP Study Guide
Module 2: AI Impacts & Responsible AI · BoK II.A

Who gets harmed: the five targets

The exam frames every harm question around who is affected. The Five harm targets are individuals, groups, society, organisations and ecosystems - and harms are not singular: one system (e.g. facial recognition) can hit several at once.

The exam frames every harm question around who is affected. Learn the five buckets and what sits in each.

The five harm targets
TargetWhat sits in it
👤 IndividualsCivil rights · economic opportunity · safety
👥 GroupsDiscrimination towards population subgroups → gender, race, ethnicity, socioeconomic status
🏛️ SocietyDemocratic process · trust in government · educational access · job redistribution
🏢 OrganisationsReputational · cultural · economic · acceleration risks
🌍 EcosystemsNatural resources · environment · supply chain
Exam flash: facial recognition

Harms are not singular - one system can hit several targets at once. The set-piece example is facial recognition → harms individuals (journalists, protesters, domestic-violence victims, wrongful identification for a crime), groups (identifying marginalised people who gather regularly at one location), society (deepfakes and spoofing eroding public trust) and organisations (privacy violations and security breaches through biometric theft).

Key terms - quick answers

What is “Five harm targets”?
Individuals, groups, society, organisations and ecosystems; one AI system can harm several at once.