As an AI researcher, I strongly believe that governments should require AI systems to undergo bias testing before deployment in public sectors. AI systems have the potential to significantly impact decision-making processes, and it is crucial to ensure that these systems are fair and unbiased. Without proper testing, we risk perpetuating and amplifying existing biases, which can lead to unfair outcomes and undermine public trust in AI technologies. Rigorous bias testing should be a fundamental part of developing and implementing AI systems, particularly those used in critical public services.