2 ms·
This just reaffirms my view is that big companies will lose headcount because of AI, but small and medium companies will (or at least have the potential to) lev
by keithnz 4mo ago
This just reaffirms my view is that big companies will lose headcount because of AI, but small and medium companies will (or at least have the potential to) leverage AI to do bigger and better things. This is because big companies could always spend big money on getting what they want made while small companies always have to tradeoff what they can realistically do with the resources they have.
- wolvoleo 4mo agoAlso big companies spend tons and tons of time and money on useless busy work. I work in IT and when we needed something new we'd just implement or build it. Now we have long certification processes for anything new, checking if it complies with hundreds of pages of policies. A lifecycle management program which we constantly have to keep updated. Governance teams that are constantly looking over our shoulders. All shit that has nothing to do with IT whatsoever. As a result we spend 90% of time doing busywork jumping through hoops these guys set up for us. Only 5% is real technical work and a lot is outsourced or consulted out to a friend of the vice president who spends all day chatting in his office for 1000 bucks a day. Or a Deloitte guy who looks great in a suit and has no idea what he's talking about. Because companies hate employing people who have actual knowledge. I really hate IT work now. Not sure about the rest of the industry but this change happened about 10 years ago. Until then we still were able to do actual useful work. I can only imagine how awful a place to work it will become when they will use AI to dream up even more inhibiting policies to keep us down with. Oh and meanwhile the CEO still goes around how innovative we are even though any innovation is absolutely killed by all this bureaucracy. Most of the time we come up with a great idea it doesn't move ahead because nobody wants to deal with years of pencil pushing to get it approved. I can totally see how startups can do actual work with little money and we can't do anything.
- georgel 4mo agoThis has been the case forever in corporate environments, even before AI. I worked for 3 years on an app that in startup land should have taken a couple months at best.
- tonyedgecombe 4mo agoThe interesting thing is how different outwardly similar businesses can be. I remember doing back to back contracts for two similar sized pharmaceutical companies. One of them had an IT group ten times the size of the other. You can guess which project overran.
- robwwilliams 4mo agoIn academic research I have seen this same trend, particularly intensely in IT and security. Lock it down, lock it up, and slow down research. A hierarchy of admins and techs taught to cover their asses.
- Macha 4mo agoIs elastic a big company? I’d put them somewhere between small and medium…
- next_xibalba 4mo ago4,000 employees (pre layoff). That’s quite large.
- dredmorbius 4mo agoStandards vary, but very roughly: - Small: < 100 employees - Mid-sized: 100 - 1,500/2,000 employees. - Large: > 1,500/2,000 employees. See: <https://www.indeed.com/career-advice/career-development/business-sizes https://www.indeed.com/career-advice/career-development/busi...> <https://learn.g2.com/business-size https://learn.g2.com/business-size>. The US Bureau of Labour Statistics identifies nine classes of businesses for employment dynamics, the largest being 1,000 employees: <https://www.bls.gov/bdm/bdmfirmsize.htm https://www.bls.gov/bdm/bdmfirmsize.htm>. I'm surprised by the latter as there are many companies with > 10k employees. The list of 100 largest US companies by headcount ends with Meta at 78,865 employees. The top ten have 309,000 or more employees, two (Walmart and Amazon) over 1 million. The top 5 are all retail, delivery, or both (Amazon). <https://stockanalysis.com/list/most-employees/ https://stockanalysis.com/list/most-employees/>