A Right to Constructive Optimization: A Public Interest Approach to Recommender Systems in the Digital Services Act

Area of Science:

• Digital Ethics
• Platform Governance
• Human-Computer Interaction

Background:

• Recommender systems on very large online platforms (VLOPs) prioritize personalization, potentially masking self-serving optimization by platform controllers.
• Existing regulations like the EU Digital Services Act have limitations in addressing the core function of these recommendation technologies.

Purpose of the Study:

• To propose a "right to constructive optimization" for individuals as democratic citizens, not just consumers.
• To advocate for recommender systems that actively promote fundamental rights and civic values within digital environments.
• To offer an alternative normative benchmark for regulating platform optimization processes.

Main Methods:

• Philosophical argumentation, drawing on Iris Marion Young's theories of justice.
• Analysis of the public function of large online platforms.
• Critique of current regulatory approaches, specifically the EU Digital Services Act.
• Development of a new normative framework: the right to constructive optimization.

Main Results:

• Large online platforms can be argued to perform a public function.
• Recommender systems should foster self-determination and self-development under conditions of inclusivity, equality, reasonableness, and publicity.
• The proposed right to constructive optimization offers a richer lens for evaluating and enriching legal initiatives.

Conclusions:

• Individuals possess a legitimate claim to influence how VLOPs optimize content and employ recommendation technologies.
• A right to constructive optimization is essential for ensuring recommender systems uphold democratic values and fundamental rights.
• This framework provides a valuable alternative to current regulatory strategies for platform governance.