Dark Pattern Detection

Definition

Dark Pattern Detection examines detecting interfaces that steer, obstruct or deceive users into choices they would not otherwise make within the broader domain of covert, deceptive or exploitative attempts to shape beliefs, behaviour, attention or collective action.

Why this matters

Dark Pattern Detection 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 detecting interfaces that steer, obstruct or deceive users into choices they would not otherwise make 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 choice reversal, comprehension and user control 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 Dark Pattern Detection

Stakeholders and beneficiaries

election authorities; civil society; platforms; journalists; security researchers; public agencies; brands; communities; targeted populations