Search papers, labs, and topics across Lattice.
This study investigates the trade-off between clinical safety and environmental impact in therapeutic large language models (LLMs) by analyzing K-Bench safety scores alongside EcoLogits life-cycle assessments across 47 model configurations. The findings reveal a significant non-linear relationship, where a modest increase in clinical safety score can lead to a drastic rise in energy consumption, with a 2.61 percentage-point increase in safety correlating to a 60-fold increase in energy use per million tokens. Additionally, the research indicates that increased computational resources do not consistently enhance safety, suggesting that larger models may not be the most effective approach for improving therapeutic AI safety.
A small boost in clinical safety can dramatically escalate energy consumption, challenging the assumption that bigger models are always better for therapeutic applications.
The deployment of large language models (LLMs) in mental health contexts raises questions about the relationship between clinical safety and environmental cost. In this paper, we examine this relationship by combining K-Bench clinical safety scores with EcoLogits life-cycle assessment estimates across 47 supported model configurations. We evaluate model performance and environmental impact across four dimensions: energy use, carbon emissions, water consumption, and abiotic depletion. The results indicate a non-linear trade-off at the upper end of the safety distribution: a 2.61 percentage-point increase in clinical safety score corresponded to an approximately 60-fold increase in estimated energy use per million output tokens. Row-level analyses further suggest that additional test-time compute did not consistently improve clinical safety and, in some configurations, was associated with lower clinical safety scores. These findings suggest that relying solely on larger models or additional inference-time computation may be an inefficient strategy for improving safety in therapeutic AI systems. We discuss the implications for sustainable deployment and highlight dynamic model selection, including model cascading, as a potential approach for reducing environmental impact while preserving clinical performance in higher-risk cases.
