Behavioral Science · AI Policy · Choice Architecture
A research publication for policymakers, enterprise leaders, and researchers who need rigorous analysis of how AI systems shape human decisions - and what governance demands it.
Read the latest intelligenceGrounded in published research on platform regulation, misinformation law, and behavioral design. About the publication →
Behavioral nudges work because they respect human autonomy while guiding better decisions. But when the nudge becomes an algorithm - when the architecture of choice is invisible and unaccountable - we cross a threshold that demands new frameworks. That is what TechNudges exists to provide.
How TechNudges designed its own AI-assisted content pipeline to preserve editorial agency - and what that design reveals about the broader challenge of building automation that respects human judgment.
TL;DR
Automation designed without explicit agency constraints defaults to efficiency, not integrity - TechNudges built explicit gates to prevent this.
The Online Safety Act shifts responsibility for online harm from users to platforms and, in doing so, shifts design authority over the choice architectures that govern what people see and share online. This piece examines three structural tensions in the Act's implementation and asks what principles should govern that authority.
Dive in →The EU AI Act classifies AI risk by application domain and potential harm. It does not systematically address the behavioural mechanisms through which AI systems influence human decision-making. This piece identifies three structural gaps and proposes behavioural impact assessment as a remedy.
Dive in →Automation designed without explicit agency constraints defaults to efficiency, not integrity - TechNudges built explici...
Nudges preserve human choice by altering its context; algorithmic systems can displace it by constructing the choice set...
Classical nudge theory requires a human designer optimising for the chooser's benefit; algorithmic feeds share the form ...
The Online Safety Act transfers design authority over online choice architectures to platforms supervised by Ofcom, with...
The EU AI Act's risk taxonomy is organised by application domain and harm category, not by the behavioural mechanism thr...
The Agency Paradox is not a failure of platform design. It is a predictable consequence of applying commercially incentivised choice architecture at scale, without the transparency or accountability safeguards that give nudge theory its ethical legitimacy.
Read the concept →