EU AI Act Officially Published, Setting New Standards for AI Regulation

The EU AI Act, a new law, comes into effect on August 1. It's significant for AI developers, but they have two years to comply. Certain provisions will take effect sooner, while others will follow later.

The law categorizes AI into three groups: low risk, high risk, and banned. Most AI falls under low risk. High-risk AI, such as biometrics and law enforcement, will face stringent regulations regarding data and bias. Banned AI includes unsettling practices like social credit scoring and widespread facial recognition.

There's a stipulation for extremely powerful AI, like GPT. They must demonstrate their processes and assess for risks. The EU aims to regulate AI without stifling innovation.

The EU's AI Office will establish codes of practice. The authorship of these codes remains somewhat unclear. The EU is seeking consultants, which has raised concerns among some. They worry the AI industry might craft its own regulations.

In essence, the EU's AI Act is a forced marriage. It's imminent, and all parties must prepare. The specifics are still being finalized, but the deadline approaches.