Yuritopiaposadism [none/use name]
technology"AI is just a tool" OK but I don't have to spend an hour explaining what a hammer is and does to the fucking hammer
EVERYTHING IS A NAIL NOW
The people who don’t work say the slop is good for workers, but the people who do work say it isn’t?
Hmmmm, who to believe?
I just had a zero productive week. So after i got tired of scrolling on my phone i just stole stuff from my boss bosses and made my own AI bot out of it. Not sure if a career move or a fuck you/me. Anyhow now my team have something useful at least. AI can be useful and no matter how much i wish the company would burn to the ground, i do see why they are the biggest fuckers around and how the tools and (some) practices could be used for good . Centralisation is good sometimes, apearantly all the libs agree when they are doing it.
work that seems polished – at least superficially – but is in fact so flawed or inaccurate that it needs to be heavily corrected, cleaned upor even completely redone
Sure it's just a simple little typo, but what timing!
This is why they call it the grauniad
a recent survey of 5,000 white-collar US workers found that 40% of non-managers say AI saves them no time at all at work, while 92% of high-level executives say it makes them more productive.
You know that output of those executives is complete ass, even in terms of AI slop
I'm not kidding when I say in my job I have a presentation I have to give weekly for senior leadership and none of them ever attend. They just read the AI summary. So basically I have to talk to a computer who summarizes it for me so an exec can read it. I asked if said meeting could be a weekly email; I can even feed my notes into an AI and just email that summary. No, it must be a spoken word presentation that the AI summarizes for leadership.
The entire fucking tech world is completely and utterly fucked in the head about AI. The day it collapses and I lose my job I'll cry tears of joy for finally being free of the madness
If they don't attend, how do they know that the input to the slopbot summarizer was in fact a spoken word presentation?
Before AI, most executives probably skimmed the summaries of written work where such summaries existed. Where summaries didn't exist they probably either didn't do the reading, sent it back to get a summary, or had to schedule some meeting with whoever produced it.
Now, they probably simplify that last step into generating an AI summary.
I can see this being interpreted as making them more productive, and in some cases that might even be true. I doubt there's an analysis of how often the AI summaries are poor quality (compared to human-written summaries). I doubt there's an analysis of how much time is wasted (and by whom) if the AI summaries mess up something important. I doubt there's an analysis comparing the benefits of having a team present its work/be questioned on its work vs. having AI simply summarize whatever they've produced.
There are also probably some superficial time savings in the form of "AI, write an email to this subordinate asking them to do XYZ." That also may save some real time in some cases, but produce externalities or extra work for other people that your executives simply don't see.
I seriously can't imagine feeling like this iteration of AI makes you dramatically more productive and not immediately succumbing to existential apocalypse.
Like, imagine if whatever it is that sustains and justifies your existence within capitalism was replaced with a dog shitting on its own feet and you and everybody else says "This is at least as good as your output if not better!" How long do you think you'd be getting credit for the dog shitting on itself? How do you bear witness to what is in front of you and not ask "What the fuck is it that I've been doing all this time?"
These people pole vault over those questions and decide that it even further justifies their privilege over the actual workers. "Yes, the commoditized and infinitely replaceable dog shitting on itself that replaces my entire productive output proves without a shadow of a doubt that I am an irreplaceable genius."
In an alternate universe this whole thing is an experiment that runs for 5 years in order to determine who is too brain broken and has to be purged from society.
I doubt there's an analysis of how much time is wasted (and by whom) if the AI summaries mess up something important.
I don't think the previous approach of skimming the documents would have uncovered the errors either. What is different now is that now not even the author-human knows what's in the documents.
Another problem is that a human author's tends to notice when something is unknown while an LLM will most of the time just fill the information gaps with plausible text strings.
I can see this being interpreted as making them more productive, and in some cases that might even be true.
Bold you to assume that executives do anything productive.
I think a large amount of skepticism is warranted, but MLs of all people should recognize that there is a real benefit from good leaders and effective organizations.
Any "leader" who thinks that this makes them more effective is not a good leader.
Under capitalism, we keep the good leaders and effective organizations right next to the unicorn stables and leprechaun pens.
Do you genuinely believe there are zero good managers or executives in capitalist countries? This is underestimating your enemy.
"Oh but it'll GET BETTER, have you tried IMPROVING YOUR PROMPTS???"
Herbert West in Reanimator after he creates another zombie with his serum, saying "not fresh enough!"
Just write "make no mistakes" at the end of your prompt ez
