Bias

Bias in AI refers to systematic prejudices or unfairness embedded in the AI's responses, often reflecting patterns present in the data it was trained on. This isn't malice, but an unintentional mirroring of societal biases found in text or images. For example, a marketing manager using AI to generate images of 'CEOs' might consistently receive images of a particular demographic if the training data was skewed. Recognizing this means you need to critically review AI outputs for fairness and actively adjust prompts to encourage diverse and equitable results.