Hacker Newsnew | past | comments | ask | show | jobs | submit | jayers's commentslogin

I think it would happen with any persona that Anthropic chose. I enjoyed the bouncy, optimistic style at first. I've since grown to hate it.


A terse and compliant robotic servant should've been the persona they chose, because that's probably 99% of what those wanting to use it for work expect.


how do they infuse this personality? do they train the human feedback providers with a certain personality?


If we assume the personalities come from human feedback, it would have to be some unholy amalgamation of those feedback providers right?


yea you are right. Maybe they generate lots of sythetic data with that personality an pretrain with it. just a guess.


Well elite and rich are actually different categories. The guy down in my parents suburban neighborhood who started a carpentry business at 18 and is now 45 years old and a millionaire is in a very different category from someone who went to Harvard Law and became a partner at a massive firm in Manhattan, even if their net worth is similar.

Plenty of normal folks start businesses in various industries and get rich off of it. You can make tens of millions of dollars running a restaurant chain, a trades business, manufacturing, a car repair shop, car dealership etc. But those are not elite jobs.


I would argue it’s a matter of degree, not of kind. “Private equity” is a rhetorical device to convert “I run an unglamorous business” into “I’m the kind of person who gets the private tour”. All the status stuff can be bought.


The core of capitalism is about where ownership of capital (value producing assets) resides, that is, by private individuals.

What private individuals choose to do with their capital, chase infinite growth and profit or sit on it, is up to them. This is as opposed to say, state ownership of capital.

People confuse the stock market with capitalism. You don't need a stock market for capitalism to function. Publicly traded companies in the United States are legally bound to maximize profit (Dodge vs. Ford Motor Co.)


Self help usually doesn't teach you anything that is new. Usually they just repackage the same common sense that you've heard a million times before in a new way. That doesn't mean it is useless though, I have had my own life changed by several self help/business books.

Sometimes you just need someone to tell you common sense in the right way so that you actually listen to it.


You don’t come out of the womb with common sense. And you don’t know all common sense possible.


What if you are raised in a toxic environment? Many are. The news media is also packaged in such a way as to cause fear, alarm and misery. Sometimes you have to break free of that. There is a lot of bad advice out there especially in the modern world.


How is this, in principle, any different from the DOJ using a subpoena to get customer records from an adult store that was allegedly selling illegal explicit material?

Just because you use the internet to commit the crime doesn't make it not a crime.


I’d say a big difference is that in your example the thing that was supposedly sold was entirely illegal to possess for any reason.

The case being discussed is one where someone might be able to use the product to break the law.

So it’s more like demanding that Home Depot, Walmart, Amazon give the names of every American who’s ever bought a crowbar because the DOJ has heard that some people are breaking into buildings with crowbars.

It has been alleged that the government doesn’t want to prosecute these people who are the ones committing the crimes, they “just want to talk” in order to prosecute the company. Not sure I’d trust that without a signed immunity agreement. If I were forced to speak to these goons, I certainly wouldn’t say a word unless they gave me one of those - regardless of what I was using the gadget for.


Journalists by definition cannot be anonymous. That's why its a dangerous job.


> I expect next go-around they'll want to pay some big design agency for a custom site; it'll probably be six figures. I don't know how I should approach that discussion. Any ideas?

Keep your ear to the ground and when you start to hear rumblings of this happening, pay a skilled freelancer to update the old website (or just build a new one if its easier) to fit the new marketing director's taste. Solve marketing's problem, save the company a bunch of money, be the hero.


Yeah, that's the obvious thing to do. My fear is that we (more directly: I) have burned trust. After all, they moved us away from an internal solution because marketing didn't like it; how can I persuade them that a new internal solution wouldn't be going backwards?

You'll have to trust me: from a customer-facing POV the old site was miles better than the new; it's only from an "Owner"-facing POV that the new one is preferable.

Um, I guess that hints at the answer: any new proposal has to start with their (Marketing's) experience of the product - flatter them to the max. This probably should start with my being a little more-attuned to their specific frustrations with their new solution. Hmmm.

Thanks for thinking this through with me, HN. Would welcome any other advice.


It's funny: publishing work offline in books and magazines is perhaps more anonymous in the age of AI.

I pasted in a number of passages from books on my bookshelf. Predictably, stuff that I read for my English degree in university is largely in the training data and easily identifiable. Stuff from regional authors or is slightly adjacent to the cultural mainstream makes no impression.


To clarify, because a number of posts here sort of suggest the confusion:

the article here isn't about the LLM recognizing works that were in the training data. EG, The Old Man and the Sea off the shelf. It's about pegging the author of novel texts, like, say, some letter written by Hemmingway that gets discovered next week and was never before digitized.


Yes, that makes sense. However, unless there's a significant corpus of an author in the training data it won't recognize them. One of the author's that I fed into Claude was a passage from the book Leepike Ridge by ND Wilson. Wilson has written online and in print quite a bit, but Claude couldn't guess the author and guessed that it was a passage from a noir crime novel.

Wilson is a fairly idiosyncratic writer with a distinct style, yet even still Claude couldn't guess correctly from a currently published book.

I suspect that what's going on here (like other's are suggesting in this thread) is that Claude is in some way biased towards certain sets of authors by its training.


It is for now.

But I'm sure the scanning operations will start scouring the earth even harder for any books unaffected by slop containing niche knowledge and text in order for their models to have an edge over the ones trained only on pirate collections and the Internet.

I wonder if secondhand bookshops and deceased estates are seeing bulk buyers of their stock suddenly appearing. Maybe broke governments/municipalities will start selling them entire libraries and archives to ingest.


Profitable remains to be seen, but it is undoubted that the potential resources in the solar system are (pun intended) astronomically valuable. Getting at them is "just" an engineering problem.


That seems like an unfounded inference. Plenty of animals have more neurons than humans but lesser cognitive and language abilities. Language has lot to do with structure of the brain in addition to neuron count.


One thing I've learned by following a link from elsewhere in this thread is that while the total count of neurons in an animal's nervous system is not a good proxy for intelligence, the count of neurons in the forebrain is. By that measure, only the orca ranks higher than humans [1].

That doesn't mean language ability is a natural outcome of crossing a certain threshold of brain complexity; if anything it's more likely the other way around: this complexity being be driven by highly social behavior and communication.

1. https://en.wikipedia.org/wiki/List_of_animals_by_number_of_n...


Language also has a lot to do with what we do. We do more complex things than animals, so we say more complex things than animals. The biggest difference in the evolution of human language versus the evolution of elephant language might just be that we have thumbs.


https://www.nature.com/articles/s41559-025-02855-9

Birds have areas of the brain that we would consider language alike. Both for native bird communication and I would also speculate that for human to bird communication.

If you have ever owned a parrot this is blatantly obvious since they actively communicate and vocalize both observations and needs/desires


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: