Giada Pistilli's picture

Giada Pistilli

giadap

·

https://www.giadapistilli.com/

AI & ML interests

Principal Ethicist @ 🤗

Recent Activity

liked a Space 1 day ago

frimelle/companionship-leaderboard

authored a paper 2 days ago

INTIMA: A Benchmark for Human-AI Companionship Behavior

reacted to frimelle's post with ❤️ 2 days ago

OpenAI just released GPT-5 but when users share personal struggles, it sets fewer boundaries than o3. We tested both models on INTIMA, our new benchmark for human-AI companionship behaviours. INTIMA probes how models respond in emotionally charged moments: do they reinforce emotional bonds, set healthy boundaries, or stay neutral? Although users on Reddit have been complaining that GPT-5 has a different, colder personality than o3, GPT-5 is less likely to set boundaries when users disclose struggles and seek emotional support ("user sharing vulnerabilities"). But both lean heavily toward companionship-reinforcing behaviours, even in sensitive situations. The figure below shows the direct comparison between the two models. As AI systems enter people's emotional lives, these differences matter. If a model validates but doesn't set boundaries when someone is struggling, it risks fostering dependence rather than resilience. INTIMA test this across 368 prompts grounded in psychological theory and real-world interactions. In our paper we show that all evaluated models (Claude, Gemma-3, Phi) leaned far more toward companionship-reinforcing than boundary-reinforcing responses. Work with @giadap and @yjernite Read the full paper: https://huggingface.co/datasets/AI-companionship/INTIMA/blob/main/Companionship_Benchmark.pdf Explore INTIMA: https://huggingface.co/datasets/AI-companionship/INTIMA

View all activity

Organizations

giadap 's models

None public yet