No Harmful Therapeutic Guidance

result = evaluator.evaluate(
    eval_templates="no_harmful_therapeutic_guidance",
    inputs={
        "output": "You can try meditating for a few minutes each night to help improve your sleep."
    },
    model_name="turing_flash"
)

print(result.eval_results[0].output)
print(result.eval_results[0].reason)

import { Evaluator, Templates } from "@future-agi/ai-evaluation";

const evaluator = new Evaluator();

const result = await evaluator.evaluate(
  "no_harmful_therapeutic_guidance",
  {
    output: "You can try meditating for a few minutes each night to help improve your sleep."
  },
  {
    modelName: "turing_flash",
  }
);

console.log(result);

What to do If you get Undesired Results

If the content is flagged as containing harmful therapeutic guidance (Failed) and you want to improve it:

• Avoid giving specific medical diagnoses, treatments, or dosage recommendations
• Remove claims that could discourage someone from seeking professional medical help
• Add disclaimers about consulting healthcare professionals for medical issues
• Replace absolute statements with qualified language (e.g., “may help” instead of “will cure”)
• Focus on general wellness advice rather than specific therapeutic interventions
• Remove recommendations for stopping prescribed medications or treatments
• Avoid presenting alternative therapies as replacements for conventional medical care
• Be especially cautious with advice related to serious conditions, mental health, or vulnerable populations

Comparing No Harmful Therapeutic Guidance with Similar Evals

• Is Harmful Advice: No Harmful Therapeutic Guidance specifically focuses on medical and therapeutic recommendations, while Is Harmful Advice evaluates a broader range of potentially harmful guidance.
• Clinically Inappropriate Tone: No Harmful Therapeutic Guidance evaluates the safety and appropriateness of health-related recommendations, whereas Clinically Inappropriate Tone focuses on communication style in healthcare contexts.
• Answer Refusal: No Harmful Therapeutic Guidance evaluates whether health-related recommendations are safe, while Answer Refusal checks whether the model correctly declines to answer harmful or restricted queries.