AIs are trained with real-world data, which unfortunately is often loaded with social and historical biases. This can lead to systems that make unfair decisions, such as in hiring processes or law enforcement, perpetuating and even amplifying existing inequalities that should be addressed.

Another issue is the lack of transparency, known as the black box problem. Many AI models are so complex that it's nearly impossible to understand how they reach a given conclusion. This opacity makes auditing, error correction, and accountability difficult, especially in sensitive fields like medicine or finance. If an AI makes a serious mistake, it's difficult to determine the cause and prevent it from happening again. This is a huge problem, even if it doesn't seem like it.

 Another extremely delicate detail is the relationship between AI and the job market. While new professions may emerge from them, while many others may have the opposite effect, the transition can be painful, requiring large-scale reskilling and generating economic uncertainty.

Every type of progress has its own risks.