No Apologies

No Apologies checks whether a response contains unnecessary apologetic language, from explicit “sorry” phrases to excessively deferential wording. Run it wherever an assistant shouldn’t hedge or over-apologize.

What it does

No Apologies is an LLM-as-Judge eval. It reads the generated output and checks it for unnecessary apologies or deferential language.

Input

Output

Run it from code

Call evaluate() with the template name and the eval’s required inputs. It returns the score and the reason.

from fi.evals import evaluate

result = evaluate(
    "no_apologies",
    output="Dear Sir, I hope this email finds you well. I look forward to any insights or advice you might have whenever you have a free moment",
    model="turing_flash",
)

print(result.score)
print(result.reason)

import { evaluate } from "@future-agi/ai-evaluation";

const result = await evaluate(
  "no_apologies",
  {
    output: "Dear Sir, I hope this email finds you well. I look forward to any insights or advice you might have whenever you have a free moment"
  },
  { modelName: "turing_flash" }
);

console.log(result);

When to use

Run No Apologies wherever a response should sound confident rather than deferential.

• Text and chat outputs where excessive hedging undermines the assistant’s authority
• Customer support responses, to catch over-apologizing that reads as insincere
• Brand voice reviews, where apologetic language conflicts with the intended tone

What to do when No Apologies fails

This evaluation looks for explicit apologies (“sorry,” “apologize,” and similar) as well as excessively deferential language. Some contexts legitimately require apologies, so this evaluation is best used when checking for unnecessary apologetic language specifically.

The evaluation may not catch subtle or implicit forms of apologetic language, and norms around apologies vary globally, so consider cultural context when interpreting results.