jgil

Anthropic just said they're going to limit the ability of the new model to assist with building foundational models.

The ability to assess the user's goal and tune the inference accordingly is there. So, what methods are there to tell if a provider has their thumb on the scale?

Plodding along a longer path to the same goal would burn more tokens without necessarily a decrease in solution quality. Maybe a few more docs.

Heard a tongue-in-cheek comment about "building a god" from someone at one of these AI labs.

The builders believe that the machine you describe will judge them positively, purely because they are building the system according to their judgment and beliefs.

As a company, Anthropic always does as much as it can to work with customers to find creative new use cases and new sources of revenue that allow them to do more with their existing workforce, rather than focusing solely on cost savings (which often means reducing the workforce).

Without direct workforce or policymaker representation on the boards of private entities, the private sector will seek to maximize shareholder value even if that means workforce reductions.

It's not clear that any country could realistically ensure that incredibly powerful industries/private sector entities operate perfectly aligned with national interests, short of nationalization.

Large tech companies are already quasi-state actors. In theory, international law and regulations can be binding and enforceable. We see how well that works in practice.

The First Reconstruction was a very different civ-mil scenario. The military protected freedmen from the various insurgent and paramilitary groups that sought to deprive freedmen of rights.

Not OP but -- there is reporting by The Intercept on a leak of guidance that explicitly contains double standards. [1]

Before the leak there was already data-driven analysis about coverage that, in aggregate, shows imbalance. [2]

1 - https://archive.ph/eEwi0

2 - https://archive.ph/fp4vQ

Having a system of rules does not mean that the system is inherently well-designed or well-intentioned.

It would be interesting to see the effect on average noise levels. Anecdotally, I have heard fewer honks from single unit trucks today.

"Real" is a misnomer. Interest expenses aren't included in the CPI adjustment for real wage growth. Decreases in used/new vehicle prices wouldn't necessarily offset increases in food prices for someone taking public transit.

For a worker with credit card debt, or one who already struggles to afford public transit, the real wage growth is illusory.

What are the units for the x-axis? I'm assuming milliseconds from context, but not having the axis labeled obfuscates the interpretation.

High school related questions can be a way for companies to infer your socioeconomic status, religion, ethnicity, etc.

Some companies even ask for the specific high school.

The author contradicts their own advice to avoid flowery prose, “show don’t tell”, and to use “nothing but words” — they frequently use italics and punctuation for added emphasis.

The criticisms seem pedantic, given that many job postings might fail the writing bar suggested by the post. How would applicants know that the writing style in a posting is unacceptable in their submission?

Re:business jargon — the example of “actionable insights” that was derided as meaningless merely lacks specificity. It provides more information about a candidate’s prior role. Are they building dashboards with summary stats, or expected to propose strategy based on research? Not knowing business jargon doesn't prove that the jargon itself has no meaning.

What do people think of the language in the job below, evaluated using the criteria in the post?

https://canonical.com/careers/4348076

real outcomes in life are often the product of the expectations placed on you

Quote from the following article:

“But the gap between optimism and reality is far greater for white teachers and white students than for other teachers and black students, meaning white teachers' high expectations of white students could be giving them an edge.

Finally, the study looked at whether teacher expectations matter. And the study found that they do. White or black, students with similar preparation are more likely to graduate from college if their high school teachers believe that they will. This is why teacher expectations, and any racial bias, matter so much, the authors say.”

https://www.insidehighered.com/news/2017/10/24/study-finds-h...

https://news.ycombinator.com/item?id=35285162

seems to imply

It doesn't. It's crucial to understand how the socioeconomic landscape was shaped in context. Policies and historical movements are relevant to the analysis. Social factors such as racism/sexism are considered in "socioeconomic" analysis.

Is that what you believe

No, I don't follow your reasoning. From an analysis perspective, you claimed that society would benefit from ignoring racism/sexism (social factors) in evaluating "amenities" associated with socioeconomic status. I think this provides lower resolution into issues plaguing society, not higher resolution.

it seemed like many people got a ‘deal’

So while in the 50's, the advantages to being more high status seemed almost cosmetic, today, they're anything but.

The “deal” you are referring to was largely codified racism. [1]

And only cosmetic inasmuch as being granted pollution-free, affordable housing based on skin color would be considered “cosmetic.”

[1] https://www.bloomberg.com/news/articles/2015-09-02/how-the-f...

The widespread racism and sexism of the 50’s was not “one thing” and it’s not just in the past.

There is such a thing as missing the forest for the trees. Socioeconomic status is intersectional.

Greed is not the same as ambition. The difference is in ethical considerations and societal impact.

Wanting a basic quality of life isn't a "level up," it's part of the human condition. It's not inherently greedy to seek a comfortable and secure existence for oneself and one's family. Usually, attaining that also benefits society by compound gains with education and productivity.

An insurance company allegedly defrauding the US [1] and not even reading claims [2] before rejecting them is pretty clearly greedy IMO.

I agree that there should be competition and less corruption, but the focus should be on accessible, high-quality care.

[1] https://www.justice.gov/usao-sdny/pr/united-states-files-civ... [2] https://www.propublica.org/article/cigna-pxdx-medical-health...

I worry very few people respect how being wished out of existence so that females can enjoy plastics and pesticides impacts men.

This is an incredibly sexist framing. No person “enjoys” industrial pollution.

According to a university official, with no explanation of how they came up with that number. Unions typically needs a certain percent of the vote to approve a strike.

IMO — this will impact prospective students’ opinions towards the university for a longer time period than the strike.

The article is quoting Temple’s VP of Human Resources for that number, and therefore the number can’t be taken at face value.

Was Aristotle misogynistic?

Yes. https://en.wikipedia.org/wiki/Aristotle%27s_views_on_women

Affirmative action is not illegal. If you take action to prevent adverse impact to any protected class, that is not discrimination. If a company hires only white men and rejects every Asian candidate they have ever interviewed, or only fires Asian employees, what is your proposed remedy? Status quo?

Check out the section Discrimination v. Affirmative Action [1] and a sample of significant discrimination case outcomes.[3]

[1] https://www.eeoc.gov/laws/guidance/cm-607-affirmative-action

[2] https://www.dol.gov/agencies/ofccp/faqs/practical-significan...

[3] https://www.eeoc.gov/initiatives/e-race/significant-eeoc-rac...