Social Signal & Information Integrity / Manipulation and Influence
SUB-T03-021Microtargeting Risks
Definition
Microtargeting Risks examines assessing harms from highly granular targeting based on inferred traits or vulnerabilities within the broader domain of covert, deceptive or exploitative attempts to shape beliefs, behaviour, attention or collective action.
Why this matters
Microtargeting Risks can materially affect autonomy, safety, public trust, market integrity, community cohesion and institutional decisions. Poorly designed interventions can suppress legitimate speech, entrench bias or create false confidence.
Research questions
Under which conditions can assessing harms from highly granular targeting based on inferred traits or vulnerabilities be measured or improved reliably, and how do effects vary by platform, population, context and intervention?
Hypotheses
A transparent, context-aware approach combining provenance, behavioural evidence and accountable human review will improve discriminatory exposure, manipulation and opacity more than single-score, content-only or opaque automated approaches.
Proposed methods
coordinated-behaviour analysis; graph analytics; campaign reconstruction; synthetic-media forensics; controlled persuasion experiments; red-team exercises; threat intelligence; expert review; affected-user interviews; reproducibility testing; methods adapted specifically to Microtargeting Risks
Stakeholders and beneficiaries
election authorities; civil society; platforms; journalists; security researchers; public agencies; brands; communities; targeted populations