Terr_ 8 hours ago

I have to give the CATO institute credit for not being co-opted, unlike some other ostensibly-libertarian outlets.

The most even-handed approach seems to be liability: What happens when an LLM generates and broadcasts severe libel? How are consumers compensated when the LLM fabricates a fake policy for non-refundable tickets?

If fixing the harms caused by using the technology in a certain way sounds too hard and too expensive, that just shows it isn't ready for those use cases.

  • brnt 8 hours ago

    > I have to give the CATO institute credit for not being co-opted

    That is because it is doing the coopting. It's why it was founded, and why it keeps being funded. "Charity" is cheaper than paying taxes, after all.

tocs3 9 hours ago

From the article:

Vance’s most concerning assertion was that:

    “[T]he Trump administration will ensure that AI systems developed in America are free from ideological bias and never restrict our citizens’ right to free speech.”
  • 1659447091 8 hours ago

    Sounds like they want to make sure the "AI systems developed in America" are free to hallucinate without consequence, including ideological "facts"