Illinois Governor Signs Nation’s Strongest Frontier AI Model Law

Posted by FallenMeringue

4 Comments

  1. FallenMeringue on

    This is relevant to the sub as it is one of the first serious state efforts to regulate AI

  2. Simultaneity_ on

    I’m not sure why we are convinced that LLMs pose a serious threat of biological, chemical or nuclear weapons. This seams like a deeply unserious conclusion by our regulators and the foundational model companies. The biggest threats seam to be in cyber security, surveillance, and IP, or illegal content generation (caught caught grok).

    Edit:
    Also in this group I would place most coding issues and bugs coming from llm generated code that gets used by people who don’t know how to review llm generated code. But then again these issues happen even with human generated code. So its a bit challenging to place the right regulation for llm designed bugs. Is it on the companies who use bad coding practices that lead to these bugs. Or is it on the model developers for false advertising these tools to these companies.

Leave A Reply