53 comments

[ 0.27 ms ] story [ 15.4 ms ] thread
"Will you lie to get this job?"

Sounds like theyre actively hiring people to train the worst kind of AI

I hate these types of questions because they're not just selecting for people dedicated to the mission, they're also selecting for people who are convincing liars
What’s the alternative? Selecting people without caring at all about their character?
I agree with the parent commenter – it's too on the nose.

If you were concerned about someone's character then you need to ask questions which really help you to understand who the person is – not questions which are easy to game by lying.

I'd argue asking something like someone's opinion on veganism is a better question because it might help you understand how the person thinks about the types of harms Anthropic are concerned about without being such an easy thing to lie about.

Obviously they could say "yeah, I think veganism is good", but someone who actually inhabited that view or was just conflicted about balancing meat consumption and the suffering of animals would probably answer with a lot more nuance that helps you understand how that person thinks about similar kinds of ethical concerns.

Fair, but I disagree. This person is interviewing at Anthropic. If they’ve done a decent job of prep work, they know something about the business, the challenges, possible adjacencies, etc.

Springing “hey what if our core business goes to zero, what then” seems qualitatively different from “what do you think of veganism” as far as learning what someone’s motives are and what their understanding of the core business is.

If I were the interviewer asking that, I would score “oh that’s fine I don’t care about money” exactly the same as “wow I’d be out of here so fast”: total failure.

I think HN folks are imagining the question, imagining the responses, and imagining the interviewers’ criteria. Which, sure, easy to decide it’s a terrible question on that basis.

Isn't mostly any question not directly verifiable subject to that? At least the more questions you ask a liar to pin them down the easier it is to eventually notice them tripping up.
I’m not sure I understand the correct answer. A public company has a fiduciary duty to the shareholders. This company opted to become public.
That's a common belief, but that duty is not absolute and does not imply the executive must choose money over morals.
> One applicant — who spoke to Axios on the condition of anonymity for fear of hurting future job opportunities — recalled being asked how they would feel if the company someday abandoned its AI ambitions for safety reasons, and that decision sent the stock to zero.

Is that a scoop? I thought that was a well known thing with Anthropic interview process. They were very transparent about it beforehand from what I remember.

On HN I’ve said things very critical of the company but that’s really not one of them, overall they had a pretty good (but intense and time consuming) interview process. I don’t understand the value of that article

I wasn't aware this was known information, but had I not, I would have assumed this was some 4D PR manoeuvre.
It’s becoming trendy to hate on Anthropic these days but their AI security record speaks for itself for years. Every AI safety benchmark/paper I’ve seen since sonnet became viable placed them far ahead of all other providers - marketing ploy or not (ala Apple “privacy” focus) the company has always had as part of their DNA thus far.
This paternalistic "safety" "DNA" is ultimately just marketing and a market capture strategy.

1) Propagandize the "dangers" of AI, stage a few incidents, talk up the risks ("it bypassed our most stringent checks! it's so smart!")

2) Build some thin veneer of "safety" and industry "self-policing."

3) Lobby for AI safety "regulation" - which you've conveniently already got an answer to! Be sure to make the regulations expensive, onerous, and impossible for foreign organizations to comply with.

4) Profit.

"anonymity, for fear of hurting future job opportunities"

Ah yes, this screams morality. We're in good hands, folks!

Nothing said that this candidate was actually hired by Anthropic, just that he or she applied.
I don't agree with some things Anthropic does but that is a cool thing.
Isn't the obvious response that they should offer a comparable comp package without any sort of stock aspect?
In my experience hype driven companies are very resistant to that even when equity/cash ratio offered is more sane.
It would be a decent interview question if such a large portion of compensation did not come from equity.

Otherwise it's basically the same as asking how someone would feel if all the sudden they lost 50% or more of their wealth. No one feels okay about that, except maybe the delusional few.

Personally I'd trade 50% of my wealth for a significantly large decrease in the chance that AI gets us all killed.
And you'd be okay with Anthropic being the ones that determine what "significantly large" means?
This reads to me like an "exploitability" metric more than anything.
This rough question has been in the culture interview since at least 2021.
Would your wife Email people to fund her company? Answer correctly and youre hired bub
Admit it HN, you have sour grapes for not having a job at Anthropic/OpenAI.
I have no clue how to get my resume past their Claude ATS so it's kind of moot.
The decision to hire me is load-bearing and genuinely yours.
I was curious until I knew someone who worked for them. I can tell you that you couldn’t pay me enough now.
Yes I want so badly to work at Antropic.

- 50% or more of my compensation locked up with what I believe has a less than 5% chance of keeping significant value

- Work with engineers who believe there job now is to vibe out a code editor, and have no idea how to fix it's many issues

- Work for a man who lives to fear monger and fain altruism

No... No I think I'd rather not actually.

The candidates should read up on effective altruism before applying to this company.

If you don’t know what that is, then you won’t understand why Anthropic are asking those sort of questions.

This is good advice. And if you don’t want to spend too much time reading about EA, imagine for yourself the inner dialogue of someone struggling with the tension between wanting to get really, really rich at any cost, and also really, really wanting other people to cherish them as a heroic-level Good Person, and also (this part is crucial) not have to sully their hands doing any actual hands-on good directly.

The ideal runs close to the output of Bill Gates’ PR machine.

Why are you so anti-anthropic? Hundreds of posts on hacker news decrying anthropic.
This is such a good example of a straight forward question that achieves the exact opposite of its purpose. What do we want? Highly ethical people that are mission focused and willing to give up on economic upside for the mission. What are we going to get? Sociopaths who are perfectly happy lying to your face about their intentions, or sycophants who are going to tell you what you want to hear.

Do you know who has the best answer to a moral quandry they've had to navigate during their career? Liars! That's a really difficult question to answer if you're honest, it's a really easy question to answer if you're dishonest.

If you really want to pursue this strategy rather than setting up bullshit interview questions you could do something much more straight forward. Agree some standard of living that's acceptable for your employees - let's say 250k in the bay area. Give them that as cash comp, and everything else goes into a "fund" which gets paid out when specific milestones are achieved, as judged by some panel. Cure cancer? You all get 10% of your deferred comp.

If you want your employees to be prioritizing things other than stock price you can literally just encode that into your compensation strategy if you like!

Obviously no one would do that, because they don't actually want to give up the comp.

Also that would open you up to lawsuits galore, since “curing cancer” is not easily definable, but share price is.
It's not difficult to define what curing cancer is, we've got tonnes of measures we can define. Let's define curing breast cancer - we have cured breast cancer if the 5 year survival rate is above 99% for all SEER stages, or less than 1 in 1000 women get diagnosed with TNM stage 1 or above. It's not that hard, it's just that deep down, they don't want to be judged on that basis.

(Details of the definition of these terms: https://www.cancer.org/cancer/types/breast-cancer/understand...)

Yeah but this is a conclusion you’d reach after a lot of studies have confirmed it and general consensus is that we’ve “cured cancer,” insofar as breast cancer, for instance, is no longer a personal or social concern at all. But testing that would take many years and there would not be a single point of inflection but rather a gradual forgetting of the illness as a cause for concern, similar to how Covid is today. So legally it would be very easy for a company to exploit this definition, claim the studies are not comprehensive enough, the statistical analysis biased, etc. etc., and really drag out the legal proceedings as long as possible and make it extremely painful for anyone to pursue their bonus. I don’t think Anthropic would do that but these sorts of things happen all the time especially when there is a lot of money involved, so tying compensation instead to something very concrete and not grounded in empirical determination, like stock valuation, makes it significantly easier. And since our society is grounded on the trade of commodities which can be given concrete valuations in exchange which are irrespective of their actual use value, the valuation of a companies stock is the only way to measure these things concretely.
Or it could be posturing. Or it could be compensation negotiation.

How does the actual Anthropic internal culture compare to that of OpenAI?

The worst is that I don't believe a single second that Amodei and his crew are not greed motivated in the end. Otherwise his company would be a lot different, more open, more open source, more available to help good cause without benefit.

What I see is more the god syndrome of persons that have are in power and have an edge. He can make the conditions so it is important for him to appear as a "divine leader", that is capable to "inspire others" with whatever is the right thing that it decides. I'm quite sure that he will not tolerate debate still.

But obviously he will not see himself like that, but will be persuaded that he is a world changer and saving the world.

Presumably you're also greed motivated, because you don't work without benefit.
The difference is that most people don't make "Look at how altruistic we are! We care™" a key component of their marketing pitch.
Earning a livelihood under the constraints of capitalism is hardly the same as hoarding wealth.
> under the constraints of capitalism

What does this mean? Was life better in soviet Russia under Communism, or as a serf in feudal England?

It means we live in a world (especially in the US) where if you aren't earning money, you die within months. Unless you are one of the tiny fraction of people who have enough wealth to live without an earned income—like the heads of all the LLM companies are.

Anyone who has enough wealth to maintain roughly their current standard of living for the rest of their life, and still works in a job with a 6+ figure compensation, is doing it out of greed.

> It means we live in a world (especially in the US) where if you aren't earning money, you die within months

What are you talking about? 11% of Americans (or rather, people in America) are on SNAP benefits[0]. And that's just food; there are lots of other benefits as well. America spends the most in the world on welfare, as a real number and as a % of GDP.

[0] https://usafacts.org/articles/how-many-people-receive-govern...

Benefits which are laborious to apply for, many who should don't qualify, structured to make you feel ashamed of having them, and getting cut repeatedly by the GOP.
Anthropic’s issue is that their concern for ethics constitutes a wall. They do not want to accept the potentiality of these tools, and somehow believe that a self-defeating stance will do less harm, even as these powers become more and more available to a larger section of the population. This inversely justifies their ownership stakes since they only benefit themselves under the guise of potential self-sacrifice that would be pathetic if it ever actually happened.

The market functions only on account of the law, which binds its activity and ensures the trustworthiness of the actors within it, and thus ethical practices are already assumed by the entities that operate in the market as the condition of their operation (since if the law is unethical than it’s also unjust and therefore illegitimate; such is the frequent judgement of those who live under it). Anthropic has committed the sin of conflating this principle with what is best for society, or humanity more broadly. What would be best for humanity is unlimited and absolute power over any and all external forces. The law, and the ethics which are presupposed within it, imposes external force on humanity—it cannot be best for humanity, demonstrably. And what is worse is, market advantage only appears when some (relatively small) increase in power is offered from some good or service, so something that fundamentally goes beyond the force of the law and perhaps even breaks it at the level of its intelligibility. The market is at odds with its own condition of possibility, it is internally contradictory, and attempts to reconcile this contradiction only lead to delusional behavior and appear fundamentally in the structure of the commodities which are produced for the sake of the market. What is worst is to thus be an idealist about the ethical nature of the market (“would you be ok with your shares going to 0 because we don’t want to be unethical”) rather than be an idealist about the potentiality of human power, which cannot be limited by the market if it seeks its own absolute expression.

How does that diminish the observation about Amodei?
I can confirm that I'm totally greed motivated! I have already turned down nice job offers because the benefits were too low.
When he goes on TV and claims the models may be conscious or may automate 90% of SWE in 6 months it is designed to pump the stock price. Its ironic they hold a fictitious scenario over candidates heads like “would you still keep working here if our revenue were $0 and we cant make payroll” (the realistic scenario where the stock price literally drops to $0)
"Damn kids today, working for their money, not mine!" Since Dario believes in the mission, he should give away all his stock and work for a regular employee salary
I don't understand the problem with this question. I could barely discern the question in the article.
Based on Opus 5, I can only conclude they've activated their plan to drive the stock price to 0
TBH I find it very out of touch to ask that kind of a question. Among regular people, everyone works for money. I consider myself a "mission" driven person but that "mission" is 4th of 5th on the list of reasons to join a company. Reason number 1 has always and will always be getting money in the door to support my family.
Why not ask the same question your wife asked Epstien? Oh they do?