Understand bias, recognize equity and access issues, and explore accessibility benefits
AI models are trained on real-world data, which can reflect societal biases depending on the source of its training information. When AI replicates these statistical patterns, there is a risk that it can unfairly apply this information (AI bias). To mitigate this, always double-check AI-generated sources, re-prompt with diverse perspectives, and diversify any examples it generates. On a positive note, you can use an LLM to promote equity and diversity by specifically prompting it to give diverse outputs, building equity in your classroom.
From an equity perspective, schools and districts can ensure all students have fair access to reliable tools and training—not just a select few. AI can be a powerful tool for accessibility as well. It can quickly translate complex texts into home languages, adjust reading levels for varied learners, and convert text into alternative formats, helping to differentiate for diverse student needs.
