A Los Angeles jury has recently awarded $6 million in damages, ruling that a social media platform’s algorithms caused a young woman to develop a severe addiction, ultimately harming her mental health.
This landmark decision highlights an urgent global reality: we desperately need healthier, more transparent Recommender Systems, alongside interventions that increase users’ online resilience. This intersection between technology and psychology is exactly what we have been researching in recent years here at the BiConnect Lab.
Our Feature on Tusk & Quill
In a recent episode of the Tusk & Quill podcast, we discussed our research on designing recommender systems that prioritize user well-being alongside traditional engagement metrics. By modeling the interaction between human decision-making and recommendation algorithms, our simulations demonstrate that it is entirely possible to balance platform engagement with people’s mental health.
Two Papers Accepted at AAMAS 2026
Building on this ongoing mission, we are thrilled to announce the acceptance of two new papers at the upcoming AAMAS 2026 conference in Cyprus. These works tackle these exact challenges, ranging from the auditing of opaque social media algorithms to the development of self-regulating social media systems.
Toward Recognizing Social Media Recommenders under Absent Recommendations
Sabrina Guidotti, Dimitri Ognibene, Giuseppe Vizzari, Udo Kruschwitz, Gregor Donabauer, Davide Taibi
Social media recommender systems strongly shape collective attention while remaining largely opaque. This work formalizes the challenge of recommender identification under missing recommendations and introduces SM-ARR-G, a first approach based on graph neural network and simulated infospheres. The contribution advances algorithm auditing and transparency, supporting evidence-based assessment of AI influence in online social environments.
→ Link to OpenReview
Mitigating Problematic Social Media Use through Paired Recommender Systems with Contrasting Objectives
Stefano Livella, Luca Bolis, Sabrina Patania, Matteo Papini, Dimitri Ognibene
This study explores how recommender systems can be designed to regulate engagement rather than maximize it indiscriminately. Through simulation, we show that pairing engagement-driven and regulation-oriented modules can reduce addiction-like behaviors while maintaining a sustainable engagement, pointing toward more responsible and value-aligned recommender architectures.
→ Link to OpenReview
Cases like the recent trial in LA clearly show that thoughtful policy, regulation, and robust scientific research will be critical to creating healthier digital environments moving forward.