Faithfulness vs. Safety in AI for Medicine скачать в хорошем качестве

Faithfulness vs. Safety in AI for Medicine

In high-stakes domains like medicine, it may be generally desirable for models to faithfully adhere to the context provided. But what happens if the context does not align with model priors or safety protocols? Our evaluation across multiple frontier LLMs reveals that in the presence of counterfactual evidence, existing models overwhelmingly accept such "evidence" at face value even when it is dangerous or implausible, and provide confident and uncaveated answers. While it may be prudent to draw a boundary between faithfulness and safety, our findings reveal that there exists no such boundary yet. Ethical AI presentation by Jessy Li, Ph.D. (College of Liberal Arts).