ren_engineer

posted a comment with context a little bit ago, seems pretty relevant to HN with regards to discussions on data privacy, tech, crypto, etc.

made possible through their national digital ID connected to government services, relevant because the US wants to roll this out to other countries - https://www.usaid.gov/diiaindc

while you can make the case this is a valuable tool during a war, the same tech could be used to effectively destroy any dissent to the government under normal circumstances

that's kind of the point, you want to eliminate companies just doing labor cost arbitrage rather than hiring globally because they actually need a unique talent they can't find in the US. If that's the case, they will happily pay the premium because that person still provides ROI even at US pay rates. That's supposed to be how H1B visas work as well

>Where is the checkpoint for taxing foreign remote employees?

company payroll/tax reporting I'd assume. If payments are going to a worker in a foreign country, throw an additional tax on it to make it so there's no cost savings compared to hiring locally. That way the only reason to do it would be because you couldn't find the talent locally

>I'll tell you that I'm getting overseas talent for roles where 10 years ago I would have hired entry level talent in the US

the US is going to once again cripple one of their critical industries for short term gains, what happens when your senior engineers retire and there are no replacements prepared because it's more efficient to hire foreign senior engineers rather than onboarding and training a college grad? Even before AI improvements and Covid increasing remote work, the bar for hiring entry engineers was absurd and most startups only wanted seniors

I can't blame companies for doing this on the micro level because it makes sense, but macro level the US government needs to take action and treat hiring remote similarly to bringing in a visa worker or add a form of tariff like you would for manufactured products. Government's job is to reign in market forces that are net bad for the country due to unaccounted externalities

Eugenics was hot among progressive academics in those days, the founder of Planned Parenthood was one and so was the guy whose statistical methods are still the basis for all modern research

back then being against eugenics meant you were a crazy religious nutcase who couldn't accept scientific fact about evolution and genetics

it's just the classic:

>person in tech spends 5 minutes looking into a subject and thinks they know more than the collective knowledge of every person in the industry

it's not about trust, they will always be indirectly controlled by the US government who could force them to leak/release your data. OpenAI is already working with the US military and removed their restrictions on allowing their AI to be used for military purposes

https://time.com/6556827/openai-us-military-cybersecurity/

The SAT was based on IQ tests used during WWI to try and find draftees with the most potential who shouldn't be wasted in the trenches. Standardized tests have always been about making things as fair as possible, and the fact that paying huge money for tutoring only results in minor score improvements should be a sign they are effective

the war against them was because people didn't like the implications that poor Asian immigrants were able to outscore other demographics regardless of income, which goes against modern academia's favored nurture over nature mindset and destroys the basis for trillions in social programs. Standardized tests effectively threaten the entire house of cards, even more than the replication crisis showing that most of the "research" used to justify modern social programs was fake in the first place

I'm not sure I'd use GRRM to support anything regarding writing productivity

>people want to live in cities

no, a ton of jobs outside the cities got outsourced or automated, leaving people with no choice. The number of people who left cities when remote work gave them the chance should make that obvious

really comes down to strong ideology of any kind that rejects or censors other opinions that go against their ideology.

The Soviets rejected Mendelian genetics and modern agriculture because it came from a monk and went against communist ideology. Lysenkoism killed 50 million+ in China and Russia as a result of the famines it created

https://en.wikipedia.org/wiki/Lysenkoism

the impact of violence vs other causes of death is similar, 9/11 killed 3K people and we went to war for 20 years and spent trillions as a result. Meanwhile 100K Americans died from opioids last year and the government does nothing

I'd say the main issue is why does it take ingesting the entire collective knowledge of humanity for LLMs to work if they are actually reasoning and understand things from first principles? Humans don't require that volume of training data to learn how to write or do basic math

all the best teachers I had in public school were people who got into teaching after having an actual career because they had actually lived in the real world. Most of the "teaching" is just stuff mandated by the government and taught via lesson plans created by the book companies.

another issue is that teachers get promoted based on years in the system or by getting more useless degrees, rather than on how well they actually teach their students

makes a lot of sense in the Southwest where the land can't be used productively for much else. They should create some programs to incentivize some of the farmers in California and other states to put up solar on their land so they stop consuming so much water as well

should be noted this research was done at the Beijing University of Chemical Technology, which means researchers were messing around with this virus in a non BSL-4 lab. Pretty much every expert opinion I've seen online says this research is insane and provides no value for the risk involved

cheaper when not including externalities(losing critical manufacturing capabilities), which is why a proper government would take action. But the problem is our government is heavily lobbied to not protect critical industries in the name of short term corporate profits. And in many cases our government is outright hostile to manufacturing by holding domestic producers to pollution/labor regulations but allowing imports from countries with no regulations

going to get an MBA without having your job pay for it or any legit plan is itself a red flag to me in terms of the quality of a candidate. Why would you hire somebody who spent 100-200k(and even more money in salary opportunity cost) for a credential?

how else would you get private or recent data into an LLM without some form of RAG? The only aspect that might not be needed is the vector database

this applies to an entire national economy, Silicon Valley was successful due to having hardware and software right next to each other, same thing for US manufacturing in general when we were a powerhouse.

We then let our government get bribed by Wall Street to allow outsourcing everything for a few decades of short term profit. Now China has the manufacturing ecosystem that allows them to innovate more efficiently

maybe those universities can use some of those multi-billion dollar endowments to pay a decent wage to their grad students and professors? Always find it funny how these institutions who are statistically dominated by left wing individuals refuse to adapt and refuse to treat their staff like humans

the ultra wealthy put all their assets into trusts or charities that dodge taxes and don't require selling anything. They can then take out loans using those assets as collateral when they need cash

it's basically impossible for US inflation to get to baseline until the federal government stops the massive deficit spending, Fed can only do so much by raising interest rates when the government is spending 25% of GDP and 1/3rd of that is deficit

>But then, looking forward years in time, people start leaning more heavily on that and stop posting to Stack Overflow and the well of information for AI to train on starts to dry up

for coding tasks I'd imagine it could be trained on the actual source code of the libraries or languages and determine proper answers for most questions. AI companies have seen success using "synthetic" data, but who knows how much it can scale and improve

I love having our energy grid being reliant on a geopolitical rival. I'm sure this won't end in disaster

>China makes 80% of global solar panels

>controls 95% of overall supply chain for solar

https://www.downtoearth.org.in/news/energy/china-to-dominate...

the Silicon valley of the last 15 years was a 0% interest rate phenomenon, going to take a long time for normalcy to return after a decade of distortion to organic market forces

you'd be amazed at how many startups waste 100s of thousands(and millions) of dollars on buying seats for tools that barely anybody uses. Interest rate increases have made VC startups get a little smarter, but a few years ago it was really bad. Similar to how tons of startups burn huge amounts of money on AWS due to laziness

the issue is it takes effort to determine if the idea is good or finding subtle errors in generated code, generating it with GPT requires almost no effort from the person who then offloads it

They also allow for creating custom actions using structured data to external APIs and file upload for RAG type use cases. And they can call out to code interpreter, DALLE, and Bing