Comment by mattw2121
10 hours ago
This is great for him until the goals given him are unachievable, in the given timelines, by human thinking and reasoning alone.
10 hours ago
This is great for him until the goals given him are unachievable, in the given timelines, by human thinking and reasoning alone.
I'm not him, but I would guess that he would respond by saying that we have a moral and ethical responsibility to remain in control of what we're putting into the world. If there's a conflict between the given timelines and how quickly a human can understand the problem and the implications of the decision being made, then the thing that needs to budge is the timeline.
> moral and ethical responsibility
We do not live in a system that values these
“Not with that attitude!”
Standing by morals and ethical responsibilities is hard, it takes mental and physical energy and there are plenty of ethical fights we all have to concede every day as a result. But at the “society” level, I think this sort of rhetoric just leads us further down the hole. We can continue to fight the battles that we have the willpower and means to fight, and that’s a very individual decision. Principled people with high tolerance for bullshit (seemingly like the guy in TFA) can make a lot of difference, IMO.
1 reply →
s/moral and ethical/professional/
This just reminds me of the whole bun fiasco. If your manager asks you to rewrite a years-old codebase in a week for immediate deployment, is it not acceptable to push back?
If you work in a regulated industry, or even just if your product has had a tough relationship with incidents of late, that makes this kind of push back even more defensible.
We live in a world where most (almost all?) humans value them.
1 reply →
He probably has a lot to say on the topic, but it’s not up to him. The accommodations only extend to reasonable affordances, but do not entirely exempt the person from requirements to perform to the level of expectation for the job.
There's nothing in the article to suggest that performance of his actual duties as an administrator have been called into question. The issue under contention, as the article frames it, is the University trying to dictate the details of how that work gets done.
There are some who would call that micro-management.
And Delaware would say, what responsibility? The state wants to sanction “AI-managed” corporations. https://oversight.house.gov/release/comer-investigates-poten...
are they "enemies of the State"?
> a human can understand the problem and the implications of the decision being made
These are weasel words that will be used to move the goalposts forever.
Guys like the one in the article will argue that we don't understand how AI works, because they don't understand how AI works.
They will never take the time to learn. If they do, they will invent a new goalpost.
I do strive to understand what I'm building, but probably not in the way you mean, or the way the luddites would demand.
The truth is, they have a result they want, and they will form new arguments indefinitely. The arguments aren't worth engaging.
Gas pedal.
I don't think that projecting opinions that you don't think are reasonable upon someone who didn't actually say those things, or clinging tightly to the Goomba fallacy with phrases like "guys like the one in the article", will get you very far toward understanding what he actually said.
Which is self-defeating. If you don't make a good faith effort to understand what he said then you're leaving yourself in a poor position to credibly argue against it.
You seem livid. Kind of bizarre.
Where is that gas pedal taking you, by the way? I can see where it’s taking people like Garry Tan: https://www.fastcompany.com/91520702/y-combinator-garry-tan-...
1 reply →
So those goals and timelines should be reconsidered; If a civil engineering firm slapped up a bridge as quickly and cheaply as possible with no concern for quality, they would cease to exist. We have no standards for software quality, and therefore anything that accelerated the perceived output is enthusiastically adopted without consideration for any future consequences.
Weinberg’s Law -- If builders built buildings the way programmers wrote programs, then the first woodpecker that came along would destroy civilization.
Exactly. I would also argue that WE DON'T NEED more software, WE DON'T NEED more features, what US, humanity, needs is MORE RELIABLE software, with less bugs, that is stable.
Big tech want to tell us that AI is great because it can write code faster than what a human can, the problem is that code is shit, is full of bugs, and is not maintainable.
Reason why my company decided to stop using AI for code that gets delivered to customers, the customer find bugs, he is not happy, you have to use time to fix bugs (for free, because it's under warranty), so we waste time, and produce code that is not easy to evolve. And even if AI is faster to write code, developers are less motivated to work, so they will just waste time on the internet while being focused on the task, and thus again, there is really no save of money.
To me all of this doesn't make sense, we want to use AI to make the work instead of US, but even if we achieve in that the result of it gives benefits only to an handful of people that are already too much rich, while the world gets polluted with all the CO2 that is produced by AI tokens, people looses jobs and gets poorer. This idea to me needs to be opposed.
We actually do have standards for some software engineering. Claude isn’t writing anything that is life/death critical.
Static analysis for example. Although I’ve heard that LLMs can now do that if a user wanted it.
All analogies are bad you can’t push reboot on a building (yes this analogy is also bad)
Imagine a civil engineer that refuses to use a computer of any kind for his work. How much do you trust scribbled long division?
Edit: what about a civil engineer that refuses to use steel. Or Arabic numerals. Or a construction worker that refuses to use power equipment. A delivery man who swears off wheels.
mattw2121's hypothetical is about timelines and quality. As long as said Civil Engineer can deliver quality work within our allocated timeline, he can do long division in binary scribbled in sand for all I care.
While I agree with you at some level, I can't help thinking of the Amish. They impose plenty of restrictions on themselves and still manage to flourish. There must be a way that is different than simply accepting "progress".
This is even more true when "progress" means fertility rates collapse. A culture that is not self sustaining will die, and far more quickly than most realize. It doesn't matter how awesome that culture is for a brief moment in time, it will cease to exist.
Amish have an overall fertility rate around 6, which is why there are now almost half a million Amish in America alone. They had less than half that in 2000. In a few decades, they'll number in the millions. If people don't start having babies then groups like the Amish will take over the world, very much turning "the meek shall inherit the Earth" into a completely fulfilled prophecy. In some ways it's perhaps even a self fulfilling one if it turns out that 'non-meek' behaviors result in unsustainable societies.
We are going to have so much high quality furniture.
The Amish are an exception, one of the rare ones (along with the handful of protected, isolated tribes). The rule is that a technologically inferior group will get rolled over by a technologically superior one for the benefit of the latter.
However will he invent the next advancement in user tracking via advertisements!
The average user here thinks they're doing far more important work than they actually are.
This is the opinion of someone with very low moral character.
At that point, humans are pretty much unemployable anyway, so why worry?
He is an educator at a public university. I don't think they push people or goals very hard.
> Until the goals given him are unachievable, in the given timelines, by human thinking and reasoning alone.
Would that not be tantamount to constructive dismissal? Consider: you may have your prayer times during the day, but they will be counted against you in performance evaluations.
> but they will be counted against you in performance evaluations.
That was my initial thought: he may get his exemption from use, but TFA doesn’t indicate that he will be evaluated differently during performance review.
To go one step further, would such allowances have to be accompanied by different performance standards, like oral exams for students with dyslexia also have increased time limits?
Or would the allowances be granted for the duration of employment, the latter being shortened due to performance problems?
Exactly. The person or people above him who wanted AI to be used against his will will ultimately achieve their goals---no matter what his beliefs are. At the end of the day, his views of humanism are futile against the force of those in favor of AI acceptance.
I can remember a time when the Borg's famous catchphrase was understood to be the kind of thing that only the bad guys would say.
I remember a time when we had to go back in time to get a whale so that machines and animals could coexist.
And I also remember that even in that case resistance was in fact not futile in the end.
[dead]