Inclusive education is a growing challenge, linked to the complexity of classroom management and student diversity. This complexity can accentuate implicit biases among teachers, particularly towards minorities, and contribute to burnout and attrition. Artificial intelligence (AI) has already shown the potential to improve and accelerate training, offering more tools to reach the goal of inclusion. However, there is still a lack of applications that directly address teachers’ implicit bias awareness and encourage genuine personal reflection within an impactful training program. Here is presented a theoretical framework proposing the use of AI, trained on standardised implicit bias tests, to create hyper-realistic, teacher-tailored simulations that provide new learning opportunities and allow teachers to self-assess their implicit biases.