With our partner Salesforce, unify sales, marketing and customer service. Accele your growth!
When AI Action Summit In Paris, the vice-president Jd vance Reaffirmed that the Trump administration would not follow the European Union in its desire to preventively regulate artificial intelligence. However, in the United States, several states take on an opposite direction. California, New York, Texas, Illinois : more than a dozen of them examine bills directly inspired by theAI ACT EUROPEespecially in the fight against algorithmic discrimination in automated decisions with high impact.
Regulation aligned with the European model
The texts under discussion resume several structuring principles of the European framework:
- Supervision of automated decisions in the critical sectors : employment, education, finance, insurance, public services and justice, where AI influences access to rights and opportunities.
- Risk management obligation : Implementation of devices to identify and correct algorithmic biases.
- Transparency requirement : AI providers will have to document their models in order to guarantee their compliance and inform user companies of potential biases.
- Strengthening sanctions and legal responsibility : in Virginia and at New-mexicnew texts establish a Private Action Lawallowing citizens to continue businesses in the event of algorithmic discrimination.
A fragmented adoption, state by state
THE Colorado paved the way with the SB 205 lawentered into force in May 2024. Its application, however, raised difficulties: the governor Jared Polis denounces an overly rigid framework and has set up a Ai Impact Task Force to specify the procedures for execution.
Other states adopt an even more restrictive approach. There California prepares reinforced legislation, while the Texas Responsible AI Gover Act (HB 1709) integrates prohibitions directly from the European model.
