Baader-Meinhof

I like that these AI idioms exist. They're like watermarks for text. It's worth the cost of humans avoiding them. Companies will eventually train their models to be undetectable, but society would be better if they didn't.

> Mark Zuckerberg overruled 18 wellbeing experts to keep beauty filters on Insta

Is there a source for this? The article doesn't seem to mention it. On that topic, should we update the title to match the article title?

To steel-man and provide a more charitable interpretation of last night:

1. Maduro stole an election. He is not legitimately in power. Many other people in power, like the military and other political factions, opposed this and wants him removed.

2. These people quietly oust Maduro in the middle of the night.

3. With the tacit approval of these folks, the US arrests Maduro for previously indicted crimes.

4. The US bombs some bases, providing plausible deniability to Venezuelan military. This was coordinated and the Venezuelans abandoned these sites ahead of time.

5. There is still stability because most of the people in charge are still there. Only the illegitimate president is gone. Venezuela can have a real election now.

Also anecdotally, I experienced the opposite. After three decades of constant fighting, the divorce was almost cathartic. Their marriage was more traumatic for me than their divorce.

Interestingly, their relationship became much better after and they are ok as friends now.

I think you have an outdated understanding of AI workflow. They generally cite their sources, which you should check, just like regular search.

I ask two types of questions:

1. Factual updates to an ongoing or recent story.

2. Analysis, e.g. "What were the economic effects of Brexit."

Without AI, I would try to read multiple opinions from different sides. But its hard for me to always know which experts to trust?

AI will present both sides, but even when AI is not hallucinating, there is still the issue of "are the experts that the AI is sourcing reliable?"

There's two parts:

1. Basic math, eg long division. This is what all the commenters are focusing on, and I largely agree with folks - we need to keep teaching this.

2. HS / College math. E.g. memorizing all the rules to solve integral calculus equations. Or solving all those circuits in Intro to EE. I agree with Wolfram here. At this point, students should be mentally developed enough to start focusing on formalizing and reducing problems rather than rote memorization or hand calculations. It's good to go over all the rules and practice a few times. But to make a whole quarter of it seems a bit much.

"At times it is only the angry who are in a position to apprehend the magnitude of some injustice. For they are the ones willing to sacrifice all their other concerns and interests so as to attend, with an almost divine focus, to some tear in the moral fabric. When I am really angry, it is not even clear to me that I can calm down—the eyes of the heart do not have eyelids—and the person making that request strikes me, to adapt a locution of Socrates’, as trying to banish me from my property, the truth. They are calling me “irrational,” but they seem not to see that there are reasons to be angry."

Agnes Callard, https://thepointmag.com/examined-life/anger-management-agnes...

I'm not disagreeing with your overall point but this reads a bit like victim blaming. "Those sexy trains were manned so skimpily, and going down those long routes all alone at night, they were just asking for it!"

High-quality writing improves information dissemination. A paper like word2vec has probably been skimmed by 10's of thousands, perhaps 100's of thousands people.

One extra day of revising is nothing, comparatively.

A lot of medicine requires $100's of millions of R&D, with a high chance of failure. No company would take that risk if their work wasn't protected. So at least in some fields, it spurs innovation.

TL;DR:

- FTX invest in a bunch of companies. Most are crypto, but he also invested $500M in Anthropic, an AI startup.

- Anthropic recently raised funds that would value FTX's equity at 3.5B - 4.5B.

- SBF wants to introduce this new valuation, as proof that he didn't squander customer money. Some people, including customers who lost money, think there would be enough money in selling the Anthropic equity to make everyone whole. SBF may think that this would absolve him of any crime.

- However, FTX is ~8B in the red. The Anthropic equity would not cover this. Also, it's not clear that FTX could get 4.5B if had to immediately sell all their Anthropic shares. Are there enough buyers at that price? Not sure.

- More importantly for SBF, the ends does not justify the means in the eye of the law. He misrepresented the riskiness of FTX by lying about FTX's relationship with Alameda. In particular, he told everyone that Alameda was treated the same as all other customers and could be liquidated when its funds got too low. In practice, SBF directed employees to allow Alameda go negative, putting customers at risk of FTX becoming unsolvent, which ended up happening.

Chrome Remote Desktop has worked for me in the past. Not very ergonomic on the phone, though.

Im curious why cities don't try to capture more of the value?

- Impose massive docking fees, cheaper cruises need to anchor far away and ferry in.

- Force passengers to buy vouchers that can only be used in that city on that date.

The supply for renters are large apartment buildings while the supply for buyers are single family homes. Some major caveats to this model, like condos. Would be interested to see condo prices during this timeframe.

This is clearly presented with a slant. I'm curious if there is a more neutral reporting on this? Like the article points to the 10x returns over 75 years as an example of the city being fleeced... But that sounds kinda like normal returns to me?

FTX supposedly had funding rounds that valued it at like $30B. This is equity, not deposits. Possibly massively overpriced due to crypto bubble, but likely that he could've extracted generational wealth from it.

There was an illuminating interview he had before being arrested:

COWEN: ...let’s say there’s a game: 51 percent, you double the Earth out somewhere else; 49 percent, it all disappears. Would you play that game? And would you keep on playing that, double or nothing? So, what’s the chance we’re left with anything? Don’t I just St. Petersburg paradox you into nonexistence?

BANKMAN-FRIED: Well, not necessarily. Maybe you St. Petersburg paradox into an enormously valuable existence. That’s the other option.

https://conversationswithtyler.com/episodes/sam-bankman-frie...

Hindsight is always 20/20 but as FTX was growing and Alameda was losing money, he could have shut down Alameda before it went negative. Then, he could've focused solely on FTX, which was apparently worth 10 of billions.

City and community colleges exist, and are mostly free. But for most people, colleges are more than just an education. College is about signaling their aptitude. And so people keep trying to pile into expensive private school for the prestige.

Any sources on this? "None at all" feels like a very bold claim when we've seen recently that a large increase to monetary supply does increase inflation.

Sure, the NHS does provide excellent care, but it's not free. The fact that the British won't come to terms with this is causing its collapse:

"In December, as many as 500 patients per week were dying in Britain because of E.R. waits, according to the Royal College of Emergency Medicine, a figure rivaling (and perhaps surpassing) the death toll from Covid-19. On average, English ambulances were taking an hour and a half to respond to stroke and heart-attack calls, compared with a target time of 18 minutes; nationwide, 10 times as many patients spent more than four hours waiting in emergency rooms as did in 2011. The waiting list for scheduled treatments recently passed seven million — more than 10 percent of the country — prompting nurses to strike. The National Health Service has been in crisis for years, but over the holidays, as wait times spiked, the crisis moved to the very center of a narrative of national decline."

https://www.nytimes.com/2023/01/25/opinion/uk-economic-decli...

Isn't the root issue here "propaganda" in ads? If so, should we focus on banning that?

Otherwise, if an ad is straight facts, even if it's making a specific argument based on your personality, is that intrinsically bad? Is it bad that we are exposed to convincing counter-arguments? Shouldn't we occasionally re-exam our beliefs, especially in light of new information?

More philosophically, should we, as free-willed people, be prevented from seeing "untruths"?

> there’s an AI pricing layer which cannot be reverse engineered so evidence of price targeting can’t be found

Is this strictly true? Couldn't we can audit the model's parameters for things analogous to the user's economic status?

Plastic bags are unfairly maligned:

- They are often quite economically produced, and only require a small amount of plastic.

- They are likely reused, often as trash can liners.

- As long as we bury them in appropriate landfills, then the carbon in the bag goes back into the ground. (See other comments on the myth of landfill space running out.)

- Manufacturing cloth tote bags has a carbon footprint 20,000 times higher than a plastic bag. https://www.nytimes.com/2021/08/24/style/cotton-totes-climat...

- Furthermore, cloth tote bags hold bacteria and are unhygienic for carrying groceries. Washing the bags cost electricity/water and probably not carbon neutral either.

Manufacturers would just transfers to the consumer. We have a CRV tax for each bottle/can here in California, and as far as I can tell, it has no affect on consumer behavior.