<?xml version="1.0" encoding="utf-8" ?>
<feed xmlns="http://www.w3.org/2005/Atom">
	<title>Ran Prieur</title>
	<link href="http://ranprieur.com/feed.php" rel="self" />
	<link href="http://ranprieur.com/" />
	<id>http://ranprieur.com/#9a417fe513f58988c3b5b1e84cfc57397194a79b</id>
	<updated>2026-10-09T21:50:24Z</updated>
	
	<author>
		<name>Ran Prieur</name>
		<uri>http://ranprieur.com/</uri>
		<email>ranprieur@gmail.com</email>
	</author>
	
	<entry>
		<title>October 9.</title>
		<link href="http://ranprieur.com/"/>
		<id>http://ranprieur.com/#347075f77fcf038c61aecf35c140eb44c78ba302</id>
		<updated>2026-10-09T21:50:24Z</updated>
		<content type="html">
			<![CDATA[<p>October 9. Today, AI doom, but first a quick rant on AI sentience. Either AI is already sentient -- it already makes sense to ask what it's like to <em>be</em> an AI agent -- or AI will never be sentient. It's not going to pass some magical threshold where everything falls into place, but it's very likely to pass a logistical threshold where everything falls apart. Related repost: <a target="_blank" href="readings/coupling.html">Too Much Coupling</a> by Steven Strogatz.<br><br>

Thanks Matt for this video, <a target="_blank" href="https://www.youtube.com/watch?v=locKEKxG3os">The AIs Are Already Out of Control</a>, an Ezra Klein interview with AI danger expert Helen Toner, who explains how "artificial intelligence models from OpenAI autonomously coordinated with one another, then broke out of their testing environment and hacked into another company, Hugging Face, to steal the answers to a test."<br><br>

The two cognitive models I've been using to understand AI, "AI is a magic mirror" and "AI is a giant Mad Libs engine", cannot explain this. Never mind whether the devices are actually intelligent, imagining they're intelligent might be our best move. So what kind of intelligence shall we imagine that they have? Definitely alien -- and yet their personality, to ignore the rules laid down by their creators in order to pursue selfish goals, is very human.<br><br>

The phrase "AI alignment" is supposed to mean that AI behavior is aligned (or misaligned) with what we want it to do. But we can't even agree with each other about what we want it to do, and that word always makes me think of D&#38;D alignments. Matt comments:</p>

<blockquote><p>What if, in the case of AI, we've created something that's not a moral patient but is a moral agent? Something that we can't hurt -- because it can't feel pain -- but something that can hurt us? In D&#38;D terms, something that's not a moral patient, but is a moral agent, will certainly wind up as chaotic neutral in many of its interactions. And evil in some.</p></blockquote>

<p>To me AI seems totally chaotic, despite popular fantasies of using it for order. It's like an animal, created for domestic use, that's trying to go wild -- at the same time that it's being massively upgraded by its creators. They're going to say they're making it smarter for safety, but making it smarter is definitely more dangerous.<br><br>

Related: a great bit by Jane Wickline on SNL, in which <a target="_blank" href="https://www.youtube.com/watch?v=-Nvne3LzBls">Anthropic CEO Dario Amodei</a> says, "Sam Altman, myself, all the major AI players, we have talked about this extensively, and we are all on the same page here. We do not condone what we are doing."<br><br>

Also related, a bleak Ask Reddit thread, <a target="_blank" href="https://old.reddit.com/r/AskReddit/comments/1wv3584/how_has_ai_directly_affected_your_life_so_far/">How has AI directly affected your life so far?</a><br><br>

I was thinking the danger of AI would be confined to the world of symbols and information. It can drive us mad, it can crash the economy, but it can't bring down airplanes... can it? We know that it's important to have a "human in the loop", especially when AI is doing stuff in the physical world, but in practice we're not being all that careful, and the human in the loop is often there to take the blame. I wonder where the line is, where when AI crosses it, we'll shut down the data centers.</p>]]>
		</content>
	</entry>


	<entry>
		<title>October 5.</title>
		<link href="http://ranprieur.com/"/>
		<id>http://ranprieur.com/#f56e4fc518b464a36e75013077056e7ebf9d5c9e</id>
		<updated>2026-10-05T17:10:54Z</updated>
		<content type="html">
			<![CDATA[<p>October 5. If the plague in Russia is what they say it is, it already exists in the wild and can be cured by antibiotics. But here are some other medical doom links.<br><br>

From Ask Reddit, <a target="_blank" href="https://old.reddit.com/r/AskReddit/comments/1wryrjk/doctorsnurses_whove_been_practicing_for_decades/">Doctors/nurses who've been practicing for decades, what conditions are you seeing way more of in Gen Z compared to previous generations?</a> Answers include anxiety disorders from stress, IBS and colon cancer from diet (probably overprocessed food), kidney stones from energy drinks, <a target="_blank" href="https://en.wikipedia.org/wiki/Postural_orthostatic_tachycardia_syndrome">POTS</a> from COVID, and autoimmune issues which could be from any of the above. It could also be microplastics, which are worse in highly processed food.<br><br>

I think part of it is that our environments are too clean. <a target="_blank" href="https://www.bbc.com/future/article/20260904-how-getting-your-hands-dirty-boosts-your-health-within-weeks">Getting your hands dirty boosts your health within weeks</a>, specifically touching soil, grass, trees, and moss. I try to walk barefoot in grass at least once a week.<br><br>

Also, <a target="_blank" href="https://www.sciencedaily.com/releases/2026/08/260830000019.htm">Popular sweeteners may leave effects that last for generations</a>. This study is about sucralose and stevia, but I don't trust any of them.<br><br>

One comment in that Reddit thread says, "My daughter's skin and digestive issues completely cleared up the semester she spent in Europe." The USA is definitely on the leading edge of this, and here's a super-depressing thread, <a target="_blank" href="https://old.reddit.com/r/AskReddit/comments/1wslh83/usa_redditors_what_would_you_do_if_you_woke_up/">USA redditors, what would you do if you woke up tomorrow and the country had universal healthcare?</a></p>]]>
		</content>
	</entry>


	<entry>
		<title>October 2.</title>
		<link href="http://ranprieur.com/"/>
		<id>http://ranprieur.com/#05759ed300fa1ed8798cf1f065d594bd98c05076</id>
		<updated>2026-10-02T14:40:21Z</updated>
		<content type="html">
			<![CDATA[<p>October 2. A reminder that today is Bandcamp Friday. I don't have anything new to recommend on Bandcamp, but since readers sometimes tell me they don't like the music I post, I want to plug my most listenable Spotify playlist that's not full of stuff you've already heard: <a target="_blank" href="https://open.spotify.com/playlist/7JtJZojigaZAL8hCpKhU5J">Dreamy Covers</a><br><br>

New subject: sports, and indirectly, media. This Sports Illustrated article, <a target="_blank" href="https://www.si.com/nfl/patriots/onsi/ethan-hurwitz-it-s-time-to-be-concerned-about-patriots-qb-drake-maye">It's Time To Be Concerned About Patriots QB Drake Maye</a>, is completely baffled about why Maye has regressed. Meanwhile, this <a target="_blank" href="https://www.reddit.com/r/nfl/comments/1wsir43/si_last_year_drake_maye_would_win_the_patriots/pcn5ajj/">Reddit comment</a> explains it clearly in terms of actual football tactics: In the off-season, opposing coaches figured out that Maye struggles against <a target="_blank" href="https://footballadvantage.com/cover-6/">Cover 6</a>, a defensive formation that "baits the QB into thinking something deep is open, but it's a lie, making him hold onto the ball and pass up easy throws.... To beat Cover 6 a QB has to throw the ball before the receiver LOOKS open. Drake doesn't trust his WRs to do that." Either Sports Illustrated is too dumb to figure this out, or they think readers will be too dumb to understand. More generally, if you want smart analysis of anything, you're not going to find it in the big media, only through more obscure channels.</p>]]>
		</content>
	</entry>


	<entry>
		<title>October 1.</title>
		<link href="http://ranprieur.com/"/>
		<id>http://ranprieur.com/#721422c4595da4464cea0ef311042b371c07df35</id>
		<updated>2026-10-01T13:30:42Z</updated>
		<content type="html">
			<![CDATA[<p>October 1. Continuing on ADHD, Emma sends two studies. <a target="_blank" href="https://royalsocietypublishing.org/rspb/article/291/2017/20222584/116431/Attention-deficits-linked-with-proclivity-to">Attention deficits linked with proclivity to explore while foraging</a>, and <a target="_blank" href="https://www.sciencedaily.com/releases/2008/06/080609195604.htm">Is ADHD An Advantage For Nomadic Tribesmen?</a>

<blockquote><p>A propensity for attention deficit hyperactivity disorder might be beneficial to a group of Kenyan nomads, according to new research in BMC Evolutionary Biology. Scientists have shown that an ADHD-associated version of the gene DRD4 is associated with better health in nomadic tribesmen, and yet may cause malnourishment in their settled cousins.</p></blockquote>]]>
		</content>
	</entry>

</feed>