November 30, 2024

Risk-based regulation and the EU AI Act

In this article:

The EU’s AI Act, published in July 2024, marks the end of a four-year legislative process, introducing the world’s first binding AI regulation, with similar legislation emerging in at least eight other countries.
These laws follow a “risk-based” approach, imposing tiered obligations on developers and deployers based on perceived threats to the public, rooted in principles of consumer protection.
The Act addresses predictable harms—economic, social, or civic—while systemic and less visible risks remain outside its scope, reflecting limitations in the regulatory framework.
Despite ambitions for principles-based governance, the AI Act focuses on finite and defined risks, leaving gaps in addressing unknown or future harms.
This regulatory vacuum allows both positive societal innovation and potential exploitation, echoing challenges seen in previous digital regulations like GDPR.
Click the link to read more details

Contact Expert Request Form