It's a fair question to ask "who are independent executive agency heads accountable to" in a constitutional context. It is true that the Executive Branch has grown far beyond what the Founding Fathers could have imagined, but the idea of a unitary executive is that the President is responsible and accountable for everything that happens in the Executive Branch. If the voters don't like what the Executive Branch is doing, they can replace the President in the next election. What happens if voters don't like what independent executive agencies are doing? There's no democratic recourse.
Think of a scenario where a President was elected with a large-ish majority and promised during the campaign to change broadband regulations to reduce broadband prices across the country. Unfortunately, the FCC commissioners were all appointed by the previous president and block this policy change that the voters clearly support. How does that square with democratic accountability?
The problem is that Congress has delegated a lot of its traditional law making power to the Executive Branch. Laws are written in vague ways with executive agencies given liberty to implement as they see fit. This gives a lot of additional power to the President (who can at least be dealt with by impeachment or being voted out in the next election) as well as independent executive agency heads (who can't be directly fired by anybody). I agree that Congress should be the ones passing laws as the excessive delegation of lawmaking by Congress is what's gotten us into the current situation
none of the authority Congress has delegated has been delegated irrevocably. if you want to change how the head of the FCC is appointed there's this thing called a law that can't be passed to change it
I work at a company trying very hard to incorporate AI into pretty much everything we do. The people pushing it tend to have little understanding of the technology, while the more experienced technical people see a huge mismatch between its advertised benefits and actual results. I have yet to see any evidence that AI is "paradigm shifting" much less "revolutionary." I would be curious to hear any data or examples you have backing those claims up.
In regards to why tech people should be skeptical of AI: technology exists solely to benefit humans in some way. Companies that employ technology should use it to benefit at least one human stakeholder group (employees, customers, shareholders, etc). So far what I have seen is that AI has reduced hiring (negatively impacting employees), created a lot of bad user interfaces (bad for customers), and cost way more money to companies than they are making off of it (bad to shareholders, at least in the long run). AI is an interesting and so far mildly useful technology that is being inflated by hype and causing a lot of damage in the process. Whether it becomes revolutionary like the Internet or falls by the wayside like NFTs and 3D TV's is unknowable at this point.
Case in point: subscription costs for everything going up to justify the "additional value" that AI is bringing _whether you use it or not_.
This would have been an additional, more expensive, subscription tier in the past.
Anecdote: Literally this morning, krisp.ai (noise cancellation software that succumbed to slop-itis two years ago and added AI notetaker and meeting summarization stuff to their product that's really difficult to turn off, which is insulting seeing how most people purchased this tool JUST FOR NOISE CANCELLING, but I digress) sent an email to their customers (me) announcing that they would no longer offer a free tier and will, instead, offer 14-day trials with all features enabled.
Why?
"As AI has become central to everyday work, we’ve seen that most people preferred the unlimited workflow once they tried it."
When I'm hiring an engineer, HR will easily let me bump up the offer by $10-20K if the candidate counters. It is nearly impossible to get that same $10-20K bump for an existing engineer that is performing extremely well. Companies themselves set up this perverse incentive structure.
This! Each time I join a new job, about 1-3 months in the door, there is a sit-down with the new line manager to check-in and give some feedback. I always talk about future compensation expectations at the time. I tell them: The market pays approximately 4-5% increase in total comp per year. That means, up 20% every 4 years. That is my expectation. If they current company is not paying that rate, I will look elsewhere for work. In almost all cases, they nod their heads in agreement. Ironically, when I come to them 3-5 years later with a new job offer in hand with a nice pay raise, 100% of them do not support matching the compensation, and view me as an un-loyal "job hopper". You just can't win with middle managers.
This is why I never do internal job transfers. The total comp doesn't change. If I do an external job change, I will get a pay rise. I say it to my peers in private: "Loyalty is for suckers; you get paid less."
Yeah, companies broke the career structure decades ago. There's no seniority rewards nor pensions to look forward to, and meanwhile companies put more budget in hiring than in promoting. They look at the high turnover rates and executives shrug. Money is being made, no changes.
It's no surprise the market adapts to the new terms and conditions. But companies simply don't care enough to focus on retention.
This has been a thing for a long time and I've thought about it quite a bit, but I still have no solutions.
I'm pretty sure it just comes down to bean-counting: "we have a new fulltime permanent asset for $100k" vs "we have a new fulltime permanent asset for $120k" is effectively the same thing, and there's a clear "spend money, acquire person" transaction going on. Meanwhile, "we spent $20k on an asset we already have" is.. a hard sell. What are you buying with that $20k exactly? 20% more hours? 20% more output? No? Then why are we spending the money?
It's certainly possible to dance around it talking about reducing risk ("there's a risk this person leaves, which will cause...") but it's bogged down in hypotheticals and kinda a hard sell. Sometimes I wonder if it wouldn't be easier to just fire staff for a week then re-hire them at a new salary.
You keep a good thing going, you buy oil for the machinery, you keep your part of the bargain and do the maintenance. You pay the correct price for the stuff you are lucky enough to have been getting on the cheap.
I like the directness of the question: "Why should I pay more when it won't burn down right this instand if I don't?" This is a question asked all over, and it is dangerous, keeping anything going requires maintenance and knowledge in how to maintain it. That goes for cars and it goes for people.
This is not business, it is miserly behaviour, it is being cheap.
The miser will find himself in a harsh, transactional, brutal world. Because that is the only way for people to protect themselves against him.
This incentive is entirely backwards. It should be "what are we losing with not spending that 20k?". You lose out on someone used to the company workflow, you waste any training you invested in them, you create a hole that strains your other 3-4 100k engineers, and you add a time strain to your managers to spend time interviewing a new member.
if you really believe you can buy all that back for 120k as if you ran short on milkk, you're missing the forest for the tree.
>Sometimes I wonder if it wouldn't be easier to just fire staff for a week then re-hire them at a new salary.
if society conditions a workforce to understand the issue, sure. But psychologically. you'd create an even lower morale workplace. Even for a week, people don't want to be dropped like a hot potato, even if you pick it up later as it cools. People want some form of stability, especially in an assumed full time role.
In my view, I have observed many good, underpaid engineers because they choose stability over higher pay. Most people are happy with slow and stead pay rises while working at the same company. Companies know this and pay accordingly. Only your top 1-10% of employees need more careful "TLC" to give higher raises and regular off-cycle feedback: "You're doing great. We are giving you a special raise for your efforts." You can mostly afford to lose the rest.
I guess that's how we got here to begin with. We take a workforce and treat is as expendable instead of as a proper team.
I suppose it will vary per industry but I can't imagine an other kind of engineering being comfortable just letting go of people mid-project because "we can afford to lose them".
There is some truth to this, but I think this way of thinking is overly simplistic. From a material standpoint, any job that can provide for you and your family's needs is "meaningful" since you can't really have a meaningful life without having basic needs provided for. From a spiritual standpoint, however, I think it is detrimental for someone to know that their job is largely pointless or achieves no tangible outcomes in the world. I think this same criticism applies to UBI and other "end of work" ideas, since a person with no job is likely to suffer from the same lack of purpose as someone who senses their job is BS. People are intrinsically wired to want to do work and make some kind of difference (even if that difference is just knowing that you helped manufacture 500 cars that day, dug a ditch that will be used for some useful purpose, whatever).
About 10 years ago I worked for Fortune 500 company that had outsourced a bunch of its technical work to offshore contractors. New CEO and CTO came in and got rid of all the contractors, replacing them with talented US-based hires. Cut the size of our tech organization down from something like 12,000 (including contractors) to maybe 3,000. Our tech generally worked a lot better after that also.
I think the first set of quotes are from people trying to sell something, while the second set are based on real data and people who have actually worked with AI. I'm a engineering manager over some infrastructure teams and we've tried to use it for network monitoring without any success. It seems like it would be handy; who doesn't want to query the state of your infra in natural language? The problem is it gives non-deterministic results and occasionally just makes things up. All the developers I know who use Github Copilot say it's very buggy and often makes coding take more time if they use it for any autogeneration.
Article II of the U.S. Constitution vests the entire executive power to the President, so technically it is the President who is responsible for following and implementing the laws that Congress has passed. Since recent Congresses (going back to at least the 70s and somewhat even to the 1930s) have written laws somewhat vaguely to give the executive branch a lot of discretion, there is a lot of legal uncertainty as to what actions are allowed in this discretion. This is why so many of Trump's executive actions are working their way through the courts as it isn't immediately clear what he's allowed to do with his executive authority vs where he is stepping on Congress's toes. For example, it is an open legal question whether the President and executive agencies are required to spend every dollar allocated by Congress or if they can decide they've already spent enough to meet the Congressional intent of the spending and can decide to not spend anymore.
I would add that no one has shown any solid business benefit or productivity gain due to AI despite there being massive incentives for someone to do so. We've tried to use AI for multiple projects within my company and none of them have panned out.
The two groups I see speaking to massive productivity gains are CEOs of AI related companies, for obvious reasons, and software engineers and influencers that want a piece of the pie or to show their bosses they’re learning the technology. I think it’s important to understand what productivity means as well. Auto generating a document isn’t a productivity gain if it contains a bunch of falsehoods or ignores data that an expert in the field might catch.
> software engineers and influencers that want a piece of the pie or to show their bosses they’re learning the technology.
This breathlessly-positive performance art shows up all the time on HN, whether or not the article is even about AI. I can believe it when someone points out specific tasks that the tool is good at and why, but when someone shows up and extolls AI as the greatest human achievement since the steam engine and how it's going to change life as we know it, it just feels like they are performing for their manager or something.
I'm not saying this isn't part of the problem, but my experience has been different. When I was in my 20s, my friends and I all lived in apartments and had parties fairly often. I recall that when I was a kid in the 90s my parents often went to small house parties as well. Now, in my 40s, neither I nor anyone I know ever goes to parties despite us all owning houses and cars and living fairly close to one another.
My theory is that people have fewer parties because people have gotten flakier about attending larger social events. It is much easier to cancel plans at the last minute with a text or a social media DM, and people always seem to want to keep their options open. We've moved to getting together only with one other couple/family at a time b/c any time we try to have larger group events half of the invite list will cancel the day of.
This kind of thing heavily depends on your doctor and insurance plan. A lot of plans won't cover things like ED meds, weight loss, or hormone therapy. TBH I'm not entirely sure that's a bad thing as long as affordable alternatives are available outside of the insurance plan. Plans covering every little thing will increase premiums, so maybe for the products that Hims sells, this is a case of the market solving a problem.
Think of a scenario where a President was elected with a large-ish majority and promised during the campaign to change broadband regulations to reduce broadband prices across the country. Unfortunately, the FCC commissioners were all appointed by the previous president and block this policy change that the voters clearly support. How does that square with democratic accountability?