If any of the polygraph operators are opposed to Patel, this would be a convenient way to get his actual loyalists fired without evidence.

But if anyone knows that already, it’s FBI staff.

I never used it in person, but the LFP (light field picture) format used by Lytro cameras was an interesting concept—you could change the focus, depth of field, and perspective after the image was captured.

Fluid construction grammar
Unscented transform
Heteroglossia
Lorenz system
Relict (biology)
Yuezhi

In order: 8 (sans sauerkraut & pickled veggies), 1, 7, 3, 5, 4, 6, 2.

It’s what happens when a very naughty function tries to divide by zero.

I think it does accurately model the part of the brain that forms predictions from observations—including predictions about what a speaker is going to say next, which lets human listeners focus on the surprising/informative parts. But with LLMs they just keep feeding it its own output as if it were a third party whose next words it’s trying to predict.

It’s like a child describing an imaginary friend, if you keep repeating “And what would your friend say after that?”

The Piper at the Gates of Dawn

Art is built on metaphor, which is an underlying connection between multiple meanings.

In semantic space, meanings are points while metaphors are vectors.

The brain generates a characteristic signal (from a sub-region of Broca’s area) when it detects grammatical errors—but it generates an identical signal when you’re listening to a grammatical sentence and need to re-parse it partway through. I think this latter case is actually the real purpose of the signal: every time it triggers, your brain is warning you that you need to stop and check the sentence again even if the meaning seems unambiguous. So the “pretending they can’t understand you” reaction could just be a reflexive response to that signal (i.e., the brain is telling them it’s confused even if there’s no logical reason it should be).

Who would even follow Elon Musk?

Candidates looking for funding.

I can’t see a significant percentage of Trump’s MAGA supporters abandoning him in favor of Musk. But I can see Musk being a divisive figure among the non-MAGA Republicans, with some refusing to work with him while others are happy to take his money.

The most likely outcome is nothing.

The second most likely outcome is a fracturing of Trump’s opposition, further weakening their ability to impede him.

If they weren’t such incompetent idiots I’d suspect it was a ploy to create a controlled opposition.

My kingdom for the lord of all horses!

Since 1995, I’ve always associated Ian McKellan with his iconic portrayal of Richard III.

Which colored my impression of Lord of the Rings a bit.

The best dog whistles come with plausible deniability included.

IMO the focus should have always been on the potential for AI to produce copyright-violating output, not on the method of training.

Why would the article’s credited authors pass up the chance to improve their own health status and health satisfaction?

Critical paragraph:

Our research highlights the importance of Germany’s unique institutional context, characterized by strong labor protections, extensive union representation, and comprehensive employment legislation. These factors, combined with Germany’s gradual adoption of AI technologies, create an environment where AI is more likely to complement rather than displace worker skills, mitigating some of the negative labor market effects observed in countries like the US.

“Happy birthday to the ground!”