This is interesting because it does not jibe at all with the people I know and how productive they are. The friends I know working big tech hybrid jobs work maybe 25 hours in a good week, whereas the friends I know in office 4/5 days a week are easily at double that. I'd imagine there are some confounding variables here. I wonder if a lot of those hybrid companies driving high revenue growth are just profit machines that don't really rely on employees being productive to grow (i.e. Google). Either way surprising findings to me!

> The friends I know working big tech hybrid jobs work maybe 25 hours in a good week, whereas the friends I know in office 4/5 days a week are easily at double that.

But are they actually more productive, or are they just spending additional hours "looking busy"?

When I worked in office I "worked" 40 hours/week because I had to. Most of that wasn't actually work, maybe about 20-30 hours of actual work, sometimes less. Working from home I have no pressure to pretend to work, there's no "butt-in-chair" requirements. I get slightly more done from home, and exchange I don't have to waste 10-15 hours/week at my desk for no reason other than to make some manager feel good about having butts in chairs.

And what about frequent chats with work friends?

No, not work, just hey how ya doin, and then someone brings coffee, then there’s a quick gab about the game, then a half baked question before a little catchup about that show. 2-4x salaries and minutes being burnt in small untracked chunks.

Hours clocked and hours worked are not the same.

When I worked at a public company I was told to expect 80% max productivity from my people. If my team routinely entered more time than that against capitalizable projects I would get called out and have to explain how that could be true.

What you are talking about is people being people and should already be accounted for. If you need to track untracked chunks and minutes something is very wrong, and it is either your management of your people or that you work at a zombie company that can't afford enough workers and is only surviving on unrealistic labor expectations.

Do they do more work by working longer hours? People often do not. Output is not the same thing as hours worked.

Productivity is output per hour, so its also possible for output to fall and productivity to rise. This will be the usual case with fewer hours.

> Output is not the same thing as hours worked.

factories would disagree. there are diminishing returns as people get tired and productivity dips, but there is still a direct, linear relationship.

however its very different from knowledge- or social-based (e.g. sales) workers. there is often no direct 1-to-1

I think you may be confusing work with "work". It's easy to find yourself spending 40 hours in an office and accomplishing 6 hours of actual work, and conversely it's easy to spend 25 hours at home and do 20 hours of actual work.

This obviously depends on what the work is. People whose job is to focus on building things for hours at a time have very different optimal work environment from people who meet with and coordinate other people in short bursts. So no one-size-fits-all top-down policy can be effective; local flexibility is required.

If someone is actually working 8 solid hours every day 5 days a week on office work, I would expect they are neurodivergent in some way. Even in college when you were still young, people couldn't sustain that level of focused intensity for very much more than maybe a week ahead of an exam, often with the help of adderall.

> don't really rely on employees being productive to grow (i.e. Google)

How do you think Google grows? They invented PageRank and have been on a set trajectory since?

Google didn't catch up to and surpass OpenAI by doing nothing.

You claim a significant amount of qualitative data here; how did you measure it? Sounds like some WAGs. Confounding variables don't matter when the "data" is guesswork.

That is a lot of thoughts and feelings. There is plenty of research you can read before throwing out opinions.

> The friends I know working big tech hybrid jobs work maybe 25 hours in a good week, whereas the friends I know in office 4/5 days a week are easily at double that.

Hours "worked" != productivity.

Do they work or is that just their at the office?

Same question applies at home. Are you working, or just logged in to Slack/Teams (or whatever) and available if it pings?

I work from home quite a bit, but I'm expected to be available and working during business hours. So other than not commuting to the office, it's not a huge difference.

If you are working from home, you are judged solely by your productivity. If you work in an office, there is productivity and there is also attendance. Someone can show up every day and appear to be working hard, yet get nothing done. You just can't do that working from home (not for very long anyways).

For most professionals, attendance shouldn't really matter. You're not creating value by merely being present. There are exceptions - customer service for example - but even in those cases being just logged into Slack/Teams is perfectly cromulent.

Exactly.

It’s rather useless to just look at the clocked hours.

The decision to buy scale and put Wang in charge was a bad one, but I think people are too focused on the $14B (which doesn't matter to Meta in the slightest) instead of the actual logic behind the purchase. Meta didn't need more labeled data, and a data labeling expert is not the person to lead your AI initiatives. I think spending $14B to try and get ahead in the AI race is a great idea, and Meta should be spending more money (there is some level of existential risk if they don't play in this market), but spending it on companies like Scale makes no sense.

This seems like the right take to me. Meta made $60 bil net profit on their last earnings report. With Mark unaccountable to investors effectively all of that can be funneled into their AI initiative. With their massive distribution network I think it makes sense for them to take big risks because if they fall behind they dont suffer a penalty from declining user base. It just never made sense for this to be the risk they took

Yeah I care about LLM's generating skills after attempting tasks and learning lessons from those attempts, not before attempting a task for the first time. This result seems a little silly and detached from the reality of how skills are "auto-generated" in the real world.

That is my approach. I don’t think the papers author has actually used skills.

Did you check our repos and sites? the repo is skills native. Also please don't be misled by the original title, we have this configuration to eliminate the impact of internal knowledge of LLMs. It's in the paper.

I disagree with most of this article. Disclaimer: I'm a junior engineer and I believe both that: 1. AI is going to take my job, 2. AI is going to do incredible good for the world.

I don't see how these are distinct. It's a technology shift, of course it's going to make certain jobs obsolete - that's how technology shifts work.

I'm not going to go through every quote I disagree with, but unlike some AI negativity discourse (some of which I agree with btw, being an optimist doesn't mean being irrational) this just reads as old man yells at cloud. Mainly because the author doesn't understand the technology, and doesn't understand the impact.

The author clearly does not understand model capabilities (seems to be in the camp that these are just "prediction machines") as they claim it's unreasonable to expect models to "develop presently impossible capabilities". This is not at all supported by prior model releases. Most, if not all, major releases have displayed new capabilities. There are a lot more misconceptions on ability, but again not going to go through all of them.

The author also doesn't understand the impact, saying stuff like "Tech doesn’t free workers; it forces them to do more in the same amount of time, for the same rate of pay or less". What? Is the author unaware of what average labor hours were like before the industrial revolution? AI is clearly going to be hugely net positive for white-collar (and with robots eventually blue-collar) workers in the near future (it already is for many).

Average labour hours in a year dramatically increased with the Industrial Revolution.

They would only decrease much later, after a long period of social conflict, economic growth, and technological progress.

During the early phase of the Industrial Revolution (roughly 1760–1850):

Agricultural workers who once labored seasonally were pushed into factory schedules of 12–16 hours per day, 6 days per week.

Annual labor hours often exceeded 3,000 hours per year per worker.

This was not because work became harder physically, but because capital-intensive machinery became expensive and had to run continuously to be profitable.

Time discipline replaced task-based work. Before industrialization, a farmer might stop when tasks were done; factory workers had fixed shifts.

This trend persisted into the late 19th century.

The labour struggle for rights we see as basic today (40h work weeks, free weekends) was bloody and deadly.

And this was without surveillance tech and automated police drones or w/e else Palantir is working on right now. If we're going by historical precedent this transition won't be pretty, even if you're hoping for a nice optimistic end result.

I'm not so sure having a beefy 401k and maybe a couple of rental properties will be enough to insulate some of the more comfortable HN posters from all the potential chaos.

I am pretty sure we are running towards a big 1929 style system correction. I may be wrong, but that people doesn't even try to contemplate this possibility seems bonkers too me. And in that case, those 401k are not going to be worth much, viz. the price of butter, and neither you can count on rental properties as a income source when everywhere became Flint, OH.

I'm not sure what your comment is trying to say.

>They would only decrease much later, after a long period of social conflict, economic growth, and technological progress.

The technological process mentioned here as Industrial Revolution, was the necessary step. It was not primarily social conflicts.

The above poster is making exactly the same point.

You are right that there was an aberration in the middle where people worked way more hours but this was when things were consolidating.

But your post implies that the preference of people was to work less hours as opposed to work more hours and earn more money, which is not the case.