Weekly Assorted Things

A periodic post of things I found interesting.

  1. Intelligence and individual differences in astrological belief.
    “Of our psychological measures we find intelligence, as measured with Wordsum, to have the largest effect size, negatively predicting belief in astrology. Education also predicts disbelief, supporting the “superficial knowledge” hypothesis. Measures of religiosity and spirituality had null effects, in contradiction of the “metaphysical uncertainty” hypothesis that a need for metaphysical beliefs causes one to believe in astrology. We find that right-wing individuals are less likely to believe in astrology, in contradiction to Theodore W. Adorno’s “authoritarian” of astrology. We also find no effect scientific trust on astrological belief. “ Oof. Via Tyler Cowen.
  2. My LLM codegen workflow atm | Harper Reed’s Blog
    Really great overview. Golden age for indie devs?
  3. New York Times goes all-in on internal AI tools | Semafor
    From actively suing to actively using. Interesting pivot. As an NYT subscriber, I look forward to seeing how it pans out.
  4. AI Essentials for Tech Executives
    Lots of useful stuff here, particularly around testing.
  5. Run LLMs on macOS using llm-mlx and Apple’s MLX framework
    Leveraging Apple silicon. I will certainly hone this a spin on my machine.
  6. The Death of Government Expertise (The Atlantic)
    “Populists are generally wary of experts, especially those who work for the government, but Musk is no man of the people: He is the richest human being in the world, and he runs major companies that rely both on government-provided expertise and significant government subsidies. As my colleague Anne Applebaum wrote, “Musk has made no attempt to professionally audit or even understand many of the programs being cut”—a willful indifference that gives away the game.”
comments powered by Disqus
Built with Hugo
Theme Stack designed by Jimmy