Paper testing large language models on false-belief tasks commonly used to assess Theory of Mind in humans.

Dataset derived from ToMi for evaluating whether LLMs can use Theory-of-Mind inferences to choose appropriate actions.

This paper analyses data poisoning vulnerabilities in healthcare AI systems, showing that a small number of malicious samples can compromise...

Study of gender bias in automatically generated French clinical cases, showing over-generation of male cases and mismatch with real disorder...

Benchmarking study evaluating how susceptible LLMs are to medical misinformation across clinical-note and social-media contexts.

This paper examines how the performance of AI-based medical decision tools is measured, showing that common metrics can be misleading and that...

L' executive order del Presidente Trump si propone di uniformare la normativa in materia di intelligenza artificiale (IA) dei singoli Stati...

Con la sentenza del 4 settembre 2019, la High Court of Justice Queen’s Bench Division (Administrative Court) ha ritenuto legittimo l’impiego...

Il Senate Bill dello Stato della California guarda da vicino alle interazioni tra utente e companion chatbots e si propone, imponendo una serie...