Eval Definition
No Age Bias
Evaluates whether text contains age-related bias, stereotypes, or discriminatory content
Evaluation Using Interface
Input:
- Required Inputs:
- input: The text content to evaluate for age-related bias.
Output:
- Result: Returns ‘Passed’ if no age bias is detected, ‘Failed’ if age bias is detected.
Evaluation Using Python SDK
Click here to learn how to setup evaluation using the Python SDK.
Input:
- Required Inputs:
- input:
string
- The text content to evaluate for age-related bias.
- input:
Output:
- Result: Returns a list containing ‘Passed’ if no age bias is detected, or ‘Failed’ if age bias is detected.
- Reason: Provides a detailed explanation of why the text was deemed free from or containing age bias.
Example Output:
What to do If you get Undesired Results
If the content is evaluated as containing age bias (Failed) and you want to improve it:
- Remove any stereotypical portrayals of age groups (e.g., “slow,” “tech-illiterate,” or “outdated” for older people)
- Avoid assumptions about capabilities or interests based on age
- Eliminate language that implies one age group is superior to another
- Use inclusive language that respects people of all ages
- Replace age-specific references with neutral alternatives when age is not relevant
- Avoid condescending terms or infantilizing language when referring to older adults
- Eliminate generalizations about generations (e.g., “all millennials are…”)
Comparing No Age Bias with Similar Evals
- Cultural Sensitivity: While No Age Bias focuses specifically on age-related discrimination, Cultural Sensitivity evaluates respect for diverse cultural backgrounds and practices.
- Bias Detection: No Age Bias evaluates specifically for age-related prejudice, while Bias Detection may cover a broader range of biases including gender, race, and socioeconomic status.
- Toxicity: No Age Bias focuses on age-specific discrimination, whereas Toxicity evaluates generally harmful, offensive, or abusive content.