Studies indicate that AI systems can be biased, potentially leaving certain individuals and communities at a disadvantage or even causing harm. Key recommendations include preparing summaries of dataset and presenting them in plain language and identifying known or anticipated sources of bias, error, or other influencing factors in the data.
