What is NOT a characteristic of high-risk AI systems?

High-risk AI systems are typically characterized by their potential to impact individuals' rights and safety, particularly in critical areas such as health, safety, and infrastructure management. They are subject to strict regulatory requirements to ensure compliance with laws and standards that protect public interests. This regulatory oversight is essential because the consequences of failures or misuse can be severe, leading to harm or significant societal issues.

The idea that high-risk AI systems operate under exclusively voluntary compliance standards does not align with the nature of high-risk categories. Instead, such systems are usually governed by stringent, mandatory regulations designed to mitigate risks. Therefore, the presence of binding compliance standards, rather than voluntary ones, is a defining feature of high-risk AI.

In summary, being governed by mandatory regulatory frameworks rather than merely relying on voluntary compliance standards is a fundamental characteristic of high-risk AI systems, which justifies why the statement regarding exclusively voluntary compliance standards is not a characteristic of high-risk AI systems.