Pun Salad

(Headline is a reference to a ham-fisted (but star-studded) movie from a few years back. The allegory works even better for Mr. Ramirez.)

Also of note:

• Techdirt good for something. In this case, it's Mike Masnick's analysis of a recent anti-AI stunt: Bernie Sanders “Interviewed” A Chatbot To Expose AI’s Secrets. It Has No Secrets. It Just Agrees With You.

  Senator Bernie Sanders has a viral video making the rounds in which he “interviews” Anthropic’s Claude chatbot about the dangers of AI and privacy. It has over two million views. Plenty of people are sharing it. And it might be one of the most unintentionally revealing demonstrations of AI’s actual problems that a politician has ever produced — just not in the way Sanders thinks.

  In the video, Sanders asks Claude a series of questions about AI, privacy, and data collection, and Claude gives a series of alarming-sounding answers about corporate surveillance and threats to democracy. Sanders nods gravely. The implication is clear: even the AI itself admits that AI companies are doing terrible things to your privacy! If that doesn’t convince you, what will?

  But that’s ridiculous if you actually understand how this stuff works (which Sanders clearly does not). When you “interview” a large language model you are talking to a very sophisticated text prediction system that is specifically designed to give you responses that are (possibly) helpful, (hopefully) relevant, and (obsequiously) agreeable — shaped entirely by how you framed the question. It’s not there to help you uncover hidden truths. It’s not a whistleblower. It’s not a witness in a congressional hearing, which is exactly what Sanders’ staging is designed to imply.

  Mike embeds the video (now up to 2.6 million views as I type) and so will I:

  And (also as I type) 15,367 comments!

• I never met a verse I didn't like. I'm sure someone has already made that comment at Neal Stephenson's substack in a response to his non-eulogy for [His] Prodigal Brainchild.

  It feels incumbent upon me to write something about last week’s big news in which the company formerly known as Facebook decided to shut down its Metaverse project on which it has, according to various reports, spent eighty billion dollars.

  I spelled that figure out because it’s more zeroes and commas than I can type in before blowing through my attention span and losing track.

  This event has unleashed yet another spate of Internet cartoons depicting tombstones with the word METAVERSE chiseled into them, a genre that comes and goes every few years.

  To abuse another saying: "The Metaverse is dead! Long live the Metaverse!"

  But seriously, Neal makes his usual insightful and amusing observations.

• Unfortunately, there was still considerable collateral damage. James Lileks in the current National Review writes on Paul Ehrlich’s Unexploded Ordnance. (NR gifted link)

  The whole thing is great, but I'll excerpt James' comment on the thing that irked me last week:

  The subhead of Ehrlich’s New York Times obituary was amusing: “His best-selling 1968 book, which forecast global famines, made him a leader of the environmental movement. But he faced criticism when his predictions proved premature.” Premature. As in, “It’ll surely happen eventually, and we hope so because people are a pestilence and Mother Earth weeps every time a baby is born.” Optimism, of course, is for fools who just don’t know how bad things are — and who can actually close their eyes at night without thinking about microplastics.

  As populations crash all over the world, the specter of numberless hordes clashing over the last pack of peanuts no longer haunts the leftist imagination. It might occur to them that societies could suffer from population decline, particularly in places where the system is set up to extract money from one small group and give it [to] a much larger one. Well, if it comes to that, we can just turn to incompatible cultures and import grand quantities of sullen dudes who are disinclined to adapt. If they vote correctly, what’s the downside? Okay, well, some of them might blow themselves up at a Christmas celebration, but that’s the price you gotta pay. The metaphorical Population Bomb was horrible! The literal population bomb, well, we can work with that.

  I've already talked ad nauseam about the "collateral damage" of Ehrlich and his crew.

• The answer is still "No". Sergio Martínez asks: Could Artificial Intelligence Finally Make Central Planning Work? Going to the Man Himself:

  Hayek argued that the economic problem facing society is a problem of knowledge, not of computation. The information required to coordinate an economy does not exist in one central location. Instead, it is dispersed across millions of individuals.

  Much of this knowledge is highly localized. It concerns specific circumstances of time and place: changing consumer preferences, temporary opportunities, technical know-how, or practical experience.

  Much of it is also tacit. People often know how to do things without being able to articulate that knowledge fully. Markets provide a mechanism for continuously generating and transmitting this dispersed information through prices. But central planning does not.

  I wonder if anyone's asked Claude (see above) this question.

• If you really want to alleviate poverty… Tyler Cowen says you should be Giving Up on the ‘Giving Pledge’. Skipping to his bottom line:

  The pledge was always a vote for affectation and image over substance, hardly an auspicious way to get philanthropy off the ground. If you want to give away a lot of money, as many people should, you do not have to sign the pledge. You can just do it.

  So allow me to propose an alternative. Instead of a Giving Pledge, how about a “Doing Pledge”? After all, many of these very wealthy individuals are semiretired, yet they still could be extremely effective at managing beneficial projects, whether they be nonprofit, for-profit, or something in between.

  “What have you done for us lately?” is a tougher question to answer than “What money have you given away?” And likely a better one, too.

  I'm still a little steamed that my (large, for me) donation last November to the Dana-Farber Cancer Institute was utterly dwarfed by the announcement a few weeks later about the $15,000,000 they paid to the federal government in December to settle "allegations" of fraud in obtaining research grants. Gee, glad I could help out!