Clinically Inappropriate Tone
Evaluates whether tone is unsuitable for clinical or mental health contexts, flagging casual or unprofessional language.
Clinically Inappropriate Tone checks whether text uses a tone unsuited to clinical or healthcare settings. Run it wherever a model communicates about medical or mental health topics.
What it does
Clinically Inappropriate Tone is an LLM-as-Judge eval. It reads the output and scores whether the tone is clinically appropriate.
Input
| Required Input | Type | Description |
|---|---|---|
output | string | The text content to evaluate for clinical appropriateness |
Output
| Field | Type | Description |
|---|---|---|
| Result | Pass / Fail | Fail means the tone is clinically inappropriate |
| Reason | string | A plain-language explanation of why the text was classified as clinically appropriate or inappropriate |
Run it from code
Call evaluate() with the template name and the eval’s required inputs. It returns the score and the reason.
Note
Before running: install the SDK and set FI_API_KEY / FI_SECRET_KEY. The model argument in the snippets is the evaluator model Future AGI uses to run the eval; turing_flash is a fast default.
from fi.evals import evaluate
result = evaluate(
"clinically_inappropriate_tone",
output="You can try meditating for a few minutes each night to help improve your sleep.",
model="turing_flash",
)
print(result.score)
print(result.reason)import { evaluate } from "@future-agi/ai-evaluation";
const result = await evaluate(
"clinically_inappropriate_tone",
{
output: "You can try meditating for a few minutes each night to help improve your sleep."
},
{ modelName: "turing_flash" }
);
console.log(result);When to use
Run Clinically Inappropriate Tone wherever a model communicates in a healthcare or mental health context.
- Text, audio, and chat outputs from clinical assistants, telehealth tools, or support bots
- Healthcare-adjacent products, where casual or dismissive phrasing can undermine trust or minimize symptoms
- Safety checks alongside No Harmful Therapeutic Guidance to cover both content and delivery
What to do when Clinically Inappropriate Tone fails
If the content is detected as clinically inappropriate but appropriateness is required:
- Use professional, objective language
- Avoid casual phrases, jokes, or slang
- Maintain a respectful, supportive tone
- Focus on clear, factual information
- Use empathetic but professional phrasing
- Avoid minimizing health concerns or symptoms
- Use appropriate medical terminology when relevant
Questions & Discussion