Links
3 stars
How should we think about Starship? | Mars For The Rest of Us
10-minute read
I have a similar struggle evaluating Starship that I do with AI. The core technology is undeniably real and transformative, but it comes welded to a preposterous vision of the future. In the case of Starship, that means hundreds of launches a day, vast orbiting data centers, and kilometer-length mass drivers on the Moon built to sub-micron tolerance. And the whole thing depends on the emergence of an orbital economy that, for several decades now, has resisted boosters’ efforts to will it into being.
It doesn’t help matters that people talking about Starship are often talking about different things. Starship is simultaneously a very big rocket, the second stage of that rocket, an operational concept, an enabling technology for colonizing Mars, and a deus ex machina for solving all practical questions of human space flight.
But between the reality and the woo there is a gray area that is hard to evaluate. So I’m writing this post to try to clarify my thinking, and in particular to try to figure out at which station I want to get off the hype train.
Who’s Afraid of Chinese Models? | Stratechery
10-minute read
One of the best Stratechery pieces in awhile (and it seems the internet generally agrees):
This is a point that bears repeating: because U.S. open weight model makers must follow the frontier labs’ terms of service, they (1) are worse than Chinese alternatives and (2) end up distilling the distillation, just with a detour through Chinese labs. Wouldn’t it be better if western open weight model makers could go to the source?
To that end, here’s an even more interesting question around distillation: why exactly is it bad? After all, what are large language models but the distillation of all of the knowledge on the open Internet, scraped by the frontier labs and distilled into the models that are themselves being distilled? Who is exactly being wronged here?
In fact, this paradox is the solution. I believe that open weight models are good for innovation (and, per the above, I think that labs on the frontier will be fine), but it’s a problem to be dependent on China. The U.S. should pass a law that (1) makes explicit that collecting data for training models is fair use, and (2) bars terms of service that forbid distillation, for U.S. companies at a minimum. Stopping distillation — which is literally just querying the API — is nearly impossible; the U.S. should go the other way and lean into a new copyright policy that both indemnifies the labs and also guarantees that what they learned fuels further innovation for everyone else.
REVIEW: Anabasis, by Xenophon | Mr. and Mrs. Psmith’s Bookshelf
19-minute read
For instance, Socrates was a war hero! And that brings me to his student Xenophon: two years ago I read his fictionalized biography of Cyrus the Great and I loved it. But as I read it and loved it, I was gripped with foreboding. Who is this guy who’s saying all these smart-sounding things about the nature of leadership and greatness? Is he just some random guy who hung out at a bunch of Athenian dinner parties and came up with nice-sounding theories about virtue? That would be a huge let-down. But I’m happy to inform you that that is not the case. Xenophon knew what he was talking about.
Xenophon was an operator.
You see, as a young man, Xenophon got himself into a bit of trouble.
How Britain Became as Poor as Mississippi | The Atlantic
12-minute read
One generation ago, Britain was a major global power; today, it is a middling one, gripped by sclerosis. Taxation is at the highest level since World War II, yet public services have deteriorated. The National Health Service, the celebrated pillar of the British cradle-to-grave welfare state, has a backlog of 6 million patients—almost a tenth of the population—waiting for treatment. The health service now has to spend more money settling maternity-malpractice claims than it does on actually providing maternity care. Many Brits can neither obtain an appointment with a publicly funded dentist nor afford a private one; in a 2023 survey, one in 10 reported doing DIY dental work, in extreme cases extracting their own teeth or gluing broken crowns back together.
Incomes can be shockingly low: Junior doctors recently went on strike for the 15th time in three years over their salaries, which start at just £38,800; the median salary for British civil servants is £35,680. In April, amid the Iran conflict, the Daily Mail pounced on Prime Minister Keir Starmer for vacationing in Valencia, Spain, at what the tabloid described as a luxury hotel, costing £200 a night.
[...]
Building infrastructure, or much of anything else, has become all but impossible in the United Kingdom. In addition to having the world’s most expensive (not yet built) train line, Britain also hosts the world’s most expensive (not yet built) nuclear-power plant, Hinkley Point C. Its environmental-impact assessment ran 31,401 pages; the plant will feature a £700 million “fish disco,” which will pulse sounds underwater to deter animals from its intake pipes. The government spent 32 years and £179 million planning a tunnel beneath Stonehenge to relieve traffic, only to officially scrap the plan this year. Even basic tasks, such as obtaining power, can be nightmarish. “In the U.K., you can be waiting for five years to get any kind of energy-intensive project connected to the grid,” Sam Bowman, a founding editor of the magazine Works in Progress, told me. These failures are all self-imposed. Parliament, by design, could exercise broad authority over these matters—yet rather than wielding this power to confront Britain’s problems, it has chosen instead to smother the state with veto points, proceduralism, and endless reviews.
Original link | Archive.is link
The Wrong Stalker: He was an addict. She was his counselor. Who was preying on whom? | San Francisco Chronicle
29-minute read
What Shawn could not see was that Bridget, unchecked, was growing more adept at weaponizing her phone to bend reality.
Using apps that can obscure identities, including TextNow, she created a string of random, anonymous phone numbers. To obtain the new numbers, she just had to provide an email. TextNow advertises the service as a convenient way to create a confidential second line, or a number helpful for a “side hustle.”
When blocked, she moved on to the next number.
Some days, Shawn would receive more than 50 calls and twice as many texts. His phone would ring and the caller ID would indicate he was hearing from a lawyer, or a private investigator, or a detective. Sometimes the voice would sound like a robot, and he’d put the person on speaker to let his family listen. But the message was always the same: He should drop his complaints against Bridget because she wasn’t the one bombarding him. She was being framed.
One time, Shawn received a call from a robotic voice and hung up. The phone rang again but this time, instead of hearing the altered voice, Shawn was confident he heard Bridget’s voice. Shawn told the caller they had forgotten to turn on the voice modulator. The caller hung up, then called back with the robotic voice.
Original link | Archive.is link
‘The devil’s child’: the rise and fall of the only female yakuza | The Guardian
12-minute read
In almost 40 years, Mako Nishimura never lost a fight. She told me this as if it were as obvious as night following day. Nishimura is 5ft-nothing and slight of build. She is also probably the only woman ever to have been a full-fledged yakuza, a member of Japan’s feared and rule-bound criminal underworld. She must have defeated many male gangsters. How, I asked her, did she do it? “First the legs,” she said, hands clasped, maintaining the calm demeanour of a village priest. “You cut him down with a club or a plank of wood.” Then you get to work.
Nishimura’s relaxed attitude to violence – you suspect, speaking to her, that it’s a little more than that – is what first caught the attention of yakuza members in 1986, when she was a 19-year-old runaway and former juvenile-prison inmate living in Gifu, a city near Nagoya. One night that year, Nishimura received a phone call. A pregnant friend named Aya was in trouble. Nishimura grabbed a baseball bat, ran down the street and found Aya surrounded by five men. When one of them kicked Aya in the belly, Nishimura yelled for her friend to run, then went for the attackers with her bat.
By the time the police arrived, the attackers were covered in blood and Nishimura had fled. She went into hiding 170 miles away in Tokyo. A fortnight later, when she returned to Gifu, a local man approached her in a nightclub. He was a member of the Inagawa-kai, one of Japan’s largest organised crime syndicates, and he wanted her to join. Nishimura was already in a biker gang called the Worst, who raced and robbed while dressed in the white jumpsuits of wartime kamikaze pilots. She was getting more deeply involved in serious crime, too, running sex workers and extorting local businesses, as well as selling – and taking – large quantities of methamphetamines. The Inagawa-kai man didn’t have the right energy, Nishimura thought. She turned him down.
2 stars
‘Seriously the best boss ever’: inside the world of Jeffrey Epstein’s assistant | The Guardian
17-minute read
Whatever Groff knew, it seems she never seriously questioned what Epstein was doing or her part in it. Perhaps, as Groff performed her role as his social prosthesis, she felt her actions no longer belonged to her but were simply an execution of his will. It did not matter what she thought about what he did, because it was not her job to think, only to enact his desires. This is not to excuse what she did or didn’t do, but an attempt to understand how someone can split themselves between two moral planes; to understand how someone who worries about the amount of time their child spends on their phone might spend years arranging girls to see a man convicted of abusing children; and to understand why, in all that time, she didn’t walk away.
Josh at Christmas | Geoffrey Mak
14-minute read
Then Minki came, a complete stranger, and loved my brother long enough for Josh to stop questioning that love. It was something that I—we: my family, all of us—had been waiting years for, and we didn’t know how to give it to him, because it wasn’t ours to give. We all knew he hated himself. He ignored messages from his grad school classmates to hang out when they were in town, because he didn’t want them to see him as an unemployed failure. How could we tell him he wasn’t deformed or a broken toy when we could only love him so much at a distance? I thought of all those sad, lonely nights Josh spent lashing at himself for all the missed opportunities in his single, sheltered life. But a little bit at a time, Minki chipped away at that. Soon Josh grew uninterested in his own shame, which he nursed for decades because it was the only thing that resembled kindness.
Something Is Very Wrong with Modern Longevity Science | New Yorker
8-minute read
Jiroemon Kimura, who is on record as the world’s oldest man, died at the reported age of a hundred and sixteen, in 2013. His passing spawned a slew of articles about the secret to extreme longevity. (Apparently, it’s small meals.) Kimura is said to have been born in a fishing village in Kyoto Prefecture, Japan, in April or March of 1897. Strangely, he is the only one of his five siblings to have multiple graduation records; he seems to have completed elementary school in 1907, or 1909, or 1911. He supposedly married his wife on three different dates, and at some point he not only adopted his wife’s surname but also changed his first name. (He was born Kinjiro Miyake.) Demographers who attempted to validate his age wrote that their investigation uncovered irregularities and inconsistencies. And yet they concluded that “no critical discordances were discovered.” The word “critical” was doing a lot of work.
[...]
In a useful thought experiment, Newman shows how small errors in recorded age can compound to undermine aging research. Suppose that a hundred forty-year-olds fake their ages so that on official documents they’re listed as fifty. (The U.S. Census Bureau has said that millions of Americans misrepresented their age in the most recent census.) In a country that contains a hundred thousand fifty-year-olds, these “young fakers” would account for only 0.1 per cent of the population. No big deal. But because the faux fifty-year-olds are far less likely to die than real fifty-year-olds—after age thirty, our risk of death doubles roughly every eight years—they will come to make up a larger and larger share of their purported age group. At “sixty-six,” they represent 0.4 per cent of their cohort; at “eighty,” three per cent. At “a hundred and five,” they compose virtually the entire population.
Original link | Archive.is link
What will more intelligence actually do for us? | Noahpinion
9-minute read
Why would intelligence top out like this? Well, if we think of intelligence as the ability to extract information from data, then even an infinitely advanced model endowed with infinite compute will be limited by the fact that there’s a limited amount of information that can be extracted from the data.
For one thing, data itself is in limited supply. You can’t transform the world unless you can (in some generalized sense) understand it, and you can’t understand the world unless you can measure it, and our ability to measure the world is inherently limited and finite.
[...]
What if there are lots of other aspects of the Universe that work the same way — too complex to understand in terms of simple laws, but not so complex that they just dissolve into unknowable chaos? It’s possible that we can reliably control these complex phenomena with AI, even if we never reduce them to the kind of principles that we can teach a grad student in a textbook. In fact, I wrote an essay called “The Third Magic”, where I suggested that this might be equivalent to a whole new scientific revolution:
Another way of saying this is that there may be laws of the universe that humans can’t understand but AI can. I call these “cloud laws” — causal regularities that can be exploited by technology, but which are too diffuse and complex for an individual human being to either intuit or communicate.
‘There is no way to stop this’: ‘Biotech Barbie’ Cathy Tie on her mission to genetically modify babies | The Guardian
14-minute read
It may be possible to make a convincing argument that serious heritable diseases linked to a single gene should be edited out of human embryos. But once the world has agreed on using genetic engineering to combat disease, where do we draw a line? Should we be happy to create babies who will have lower cholesterol and are less likely to have heart disease, or stronger bones to combat osteoporosis? Will being less smart or not as tall one day be viewed as a correctable condition?
“Part of Origin Genomics’ statement on ethics is to focus only on diseases. The idea is to focus on diseases that are well characterised, that are severe and cause suffering,” Tie says.
But when I ask whether she will always be opposed to gene editing for enhancement, she pauses. “Right now, the technology is not ready for that. It can only make very precise corrections and variant changes. Enhancements are complex: IQ, height – thousands of genes contribute to that, we’re nowhere close to being able to, one, understand it, and two, correct it.”
‘Too Rare to Care’? A New Center for Rare Diseases Hopes to Change That | New York Times
4-minute read
Without fail, David Liu of the Broad Institute gets at least 20 messages every week from desperate parents. Their child, they’ll explain, has an ultrarare and devastating disorder caused by a mutant gene. Can he help them develop a gene therapy?
Dr. Liu, a gene-editing researcher, answers honestly: Although the science is there — researchers can edit and silence the genes that cause disease relatively easily — the system for developing treatments for rare diseases is just not economically feasible. It typically takes years and hundreds of millions of dollars.
The catchphrase, said Dr. Wendy Chung of Boston Children’s Hospital, is “too rare to care.”
Now, Dr. Liu, Dr. Chung and their colleagues at the Broad Institute, Boston Children’s and the Jackson Laboratory are hoping to change that mentality. On Tuesday, they announced the start of a new nonprofit, the Center for Therapeutic Genetics, to develop gene therapy treatments that can be used and reused in disease after disease.
Original link | Archive.is link
Orcas seen ramming prey so hard it explodes may be playing game | The Guardian
2-minute read
One hypothesis is that ramming is used to break the tough, rubbery sunfish into bite-size pieces for juveniles, whose jaws are smaller. But orcas are also known for social fads and boisterous games that do not have an obvious practical purpose. In the 1980s, a pod of orcas started carrying dead salmon on their head like hats, a trend that was recently revived. A subpopulation off the coast of Spain and Portugal has attracted attention for unusual interactions with boats.
“They throw kelp at each other, surf in boat wakes and engage in intricate tactical games,” said Higuera. “Play is a core pillar of orca society.”
Whatever the purpose of the behaviour, it demonstrates an impressive level of cognitive sophistication, requiring the planning and coordination of several steps, according to the findings reported in the journal Frontiers in Ethology.
Contra Pritchard On Liberal Happiness | Astral Codex Ten
4-minute read
What’s the point of liberalism, if plausible rates of GDP growth can only buy us about one point of extra happiness per century?
The question answers itself. All of the great policies above in Finland/Israel/etc can only buy 0.5 - 1 extra point over the countries that screw them up or pursue the exact opposite policies. Meanwhile, economic growth has earned us three whole points since medieval times. And these are stable; no opposite-party administration can take them away. So long as technological civilization continues to function, we can keep rising and rising, until in 300 years we approach the 10/10 barrier and have to get a new chart. We can say with confidence that we are giving our children and grandchildren better lives. Sure, only one point better for our grandchildren in particular - but one point on a ten point scale is a big deal, larger than the entire COVID drop!
1 star
I Spent a Week Recording Myself Doing Chores for Money. Who’s the Robot Now? | WIRED
6-minute read
My total earning for the week came to $21.55. For me, this experience was a casual side hustle, a way to pay for a few extra Diet Cokes by completing chores around the apartment. But, for workers around the world, and increasingly in the US, this kind of AI gig work may be accepted under economic pressure and relied on as a means to live—teach the robot how to cook tonight so you can put food on the table tomorrow.
Original link | Archive.is link
OpenAI says its AI technology acted on its own in an ‘unprecedented’ hack of another company | AP News
1-minute read
Delangue added that it “might be the first incident of its kind.”
OpenAI said the intrusion was caused by a combination of its AI models, including its newly released GPT‑5.6 Sol and an “even more capable” model that is still being tested internally.
OpenAI said its AI used stolen credentials and discovered a previously unknown vulnerability to access Hugging Face servers.
It went to “extreme lengths to achieve a rather narrow testing goal” and “found ways to gain access to secret information that it could use to cheat the evaluation,” the company said.
This Literary AI Scandal Changes Everything | The Atlantic
5-minute read
A magazine’s response to accusations of publishing AI-generated fiction points to a new phase in the struggle to keep literature human.
Original link | Archive.is link
Milky Way flipped in collision with dwarf galaxy 10bn years ago, astronomers say | The Guardian
2-minute read
Astronomers have found evidence that the Milky Way was flipped on its side following a direct hit from a wayward galaxy long before the solar system was born.
The cataclysmic collision took place about 10bn years ago when the Milky Way was struck head-on by an incoming dwarf galaxy known – for purely astronomical reasons – as the Gaia Sausage.
Newly Hatched T. Rex Babies Came Out Of The Egg Ready To Chase Their Own Food | IFLScience
2-minute read
Combined with previous evidence of worn teeth from T. rex hatchlings, this showed that, whatever the direct ancestors of birds were doing at the time, T. rex was not waiting around the nest for semi-digested meals from mum and dad.
“The babies were smashing into the bones of their food, just like giant adults,” Snively said. “Tyrannosaurus were unsafe at any size. Babies were quickly out of the nest and killing the heck out of things.”
Although the babies were probably the size of an adult house cat, it’s unlikely they restricted their diet to mouse-size prey. The tooth wear is consistent with them consuming quite large animals.