At first I ignored that behavior, but now I respond asking them to refrain, it’s ok to tell me they don’t know or to tell me a hint they might know even if they don’t know for sure, but I have access to the same LLMs they have and I don’t need them to help with that. I tell them to just send me what they would have sent to the prompt, and if I need the LLM, I’ll do it.
It’s especially bad when they have zero idea and try to fake it with LLM.
To be fair, before GenAI a lot of them managed to manually vomit a lot of useless garbage and I ignored them then too… Of course, now they are more empowered to waste my time than before…
Especially obnoxious when they ask for your help in the first place and keep pestering you with GenAI bullshit saying they are ‘helping’ you figure out how to do whatever it is they couldn’t figure out with GenAI help in the first place…
Oh absolutely, and on all sorts of communication.
I sat befuddled at what the hell I was reading when I went to read a reply to an email I had fired off at the end of the day and didn’t check until next morning. Turned out he wanted to be ‘cute’ and so he had AI slop up his out of office reply. So instead of “Out of office, will be back on MM/DD” I had read two paragraphs of bizarre fan fiction looking junk apparently intended to make his three day weekend sound like some fantasy epic.
Someone asked me to help with something and I said sure, and then within they minute, they pasted a wall of text saying “this is what ChatGPT had to say, so it should help you get going”. It didn’t work for them, so why they thought I would both find it useful and couldn’t have asked an LLM myself, no idea. It was totally wrong and that’s why they screwed it up harder, but still thought the AI must have been right.
I’ve received multiple bug reports and pull requests with confusing long winded essays that bothered to bring up details that were both not relevant and yet also incorrect.
A celebrity graduation speaker that actually has something thoughtful to say that is realistic, relevant, and inspirational to the graduates? That’s crazy, everyone knows celebrity graduation speakers are supposed to be self absorbed and prattling on about the stuff that excites them without connecting to the graduates, except maybe to be dismissive of the graduates and look down on them.
The Wozniak era of Apple was pretty much all upside. Whatever less than nice things to say at Apple almost certainly came about more than a decade after he had left them.
They will hit you with and underpayment penalty, and they’ll actually get a pretty decent amount more than they would have otherwise gotten from you…
Yes, because the democrats are the ones to blame for why it will never ever get to a senate conviction…
Of course, this argument is a pretty piss poor one when the example held up is Musk…
His first jolt was selling a website to a relatively clueless Compaq that did nothing with it except throw money at Elon. No idea if Elon’s site was any good or not, but it didn’t matter because the reality is that it folded without an enduring impact as part of the dot-com collapse. He used his winnings to try to make X.com the first time, and was a comparative failure next to Paypal, which, somehow, agreed to merge and put him in charge and he almost tanked it. Then ebay bought it out and Musk got maddening amount of money for doing nothing but screw it up.
Tesla is probably the first example of him not actively screwing it up, though his drama around “I wanna call myself a founder” was dumb. That said, any investor could have done it, so his ‘value’ was his lottery-like wins leading to that point.
There are others that are arguable, though recently had it happen where someone kept giving out names of impressive and seemingly valuable ‘billionaires’, and we kept checking and every last one were ‘only’ millionaires. So it seems like ‘billionaire’ remains a stupidly over the top concept that isn’t particularly redeemable, with the defensibly decent folks staying under a billion through not being super greedy and/or philanthropy.
with the correct power input
That’s quite the tall order for a lot of the now ‘bread and butter’ of this bubble. Some of these systems are needing over 25 kilowatts now, being fed by 8 C19 connections.
And that’s ignoring the ones that just dump a crapton over a 48VDC bus bar and don’t have anything even vaguely approaching decent residential electrical hookup friendly.
While most will bemoan that the systems will be too loud and the gpus not really interesting for gaming, the sheer power supply issue is beyond what we’ve seen in datacenters historically. Usually the most egregious was still a few kilowatts over C19, which awkward but doable in residential in certain circumstances. Now we are talking dedicating half your residential service to breakers and outlets to feed the servers.
Neuralink I suppose is an example where his appetite for crazy risk might pay off or be terrible.
FSD implies no one else is trying for self driving, and that’s no shortage there for that ambition.
Humanoid robots at scale, they haven’t displayed a capability to suggest they have any better shot than others, and there’s plenty of less loud companies trying. None are particularly close, so Elon stands relatively alone in proclaiming success without the proof.
I don’t know that his crazy appetite for risk is on display with SpaceX. I would argue most of their real successes were relatively lower risk, and generally when the companies indulge musk at his craziest are like the other comment says, comparative failures.
Daniel Jackson
He may die a fair bit, but he won’t stay dead for long.
The plague of work chats now:
Here’s what ChatGPT/copilot had to say:
People can ask for themselves, you answering that way adds no value. Just say you don’t know.
In group chats, keep your mouth shut and let people that actually know answer. Don’t drown out the actual expert answers.
And holy hell the ones that will die on the hill that they are right because chatgpt agreed with them even when they are totally wrong…
Either glory seeking or legitimately trying to help.
Had several people feel like while they don’t know how to help, they could at least expend their tokens…
The igpu isn’t going to keep up with that much credibly anyway.
Though the chance someone other than you dies from you taking a shit is pretty far fetched, and I’m not liking having to try to conceive of how that happens.
But a lot of activities are this way. Getting on a ladder in public could kill someone, just breathing around other people could kill someone, etc etc.
But what about monster HDMI cables?
Well, driving to work is basically that. Non zero chance someone dies.
Well, for one, it seems to be an appropriate place. They speak of essays out of nowhere in slack, this isn’t slack, it’s a site. Besides, the target audience (people sending slop grenades) obviously value verbosity as a virtue in and of itself, so it may help them.
For another, LLM essays are their own annoying beast. The material contributed by the human was contained in a terse prompt. In a slack, that’s everything I wanted, and the LLM just adds fluff and buries the meat in verbosity. Even “I don’t know” is much more valuable than wasting my time with an LLM essay hemming and hawing without any more clarity than we started.