i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs.
-
@indigoparadox @nicuveo He says he is pro-breathalyzers in cars: "I think that if someone is convicted of driving under the influence, or something close to that, it is legitimate to attach a sensor-driven kill switch to stop per from driving while inebriated.", but the fact he links to a campaign against such things is pretty confusing.
@mrotteveel isn't he saying there that he is only for breathalysers if someone has been convicted before, but not everyone by default?
-
@mrotteveel isn't he saying there that he is only for breathalysers if someone has been convicted before, but not everyone by default?
@lizbian Probably, but that is not how it reads at first glance.
-
i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs. fuck that! ethical concerns front and fucking center! it is very revealing that tech is currently in such a state that the quiet part can be said out loud without any pushback.
@nicuveo I am increasingly convinced Anthropic are engaged in an intentional effort to bastardise the very concept of ethics when discussing AI so they can kill it and wear it's skin the way Murray Rothbard did to Libertarianism.
'No need to worry about the ethics of AI, I read an interview with a person from Anthropic and they have 3 highly ethical people discussing the ethics of AI, working in an ethical soundproof broom cupboard so it's all covered!'
-
@indigoparadox @nicuveo Seems like he is a quantum state where he is both for and against breathalyzers in cars.
@mrotteveel Schrodinger's toot.
-
@nicuveo "Ethical concerns aside". What does AI even do other than produce slop and make people reliant on technology. In like 90% of cases it is a net negative or does no good. That means when we add ethical concerns BACK in we have lost the plot and Gen AI should die in a hole and burn in the underworld.
What does AI even do other than produce slop and make people reliant on technology
It allows a blind creature, who has to operate a Linux installation without working speech synthesis (because the maintainers messed that up and didn't care about accessibility), even attempt to debug the mess in the first place and have an operational, accessible machine:
Want to debug it?
You can’t—because you can’t hear anything.
So you grab your phone, take a picture of the screen, feed it to an image captioning AI, and hope it tells you whether the error dialog says “Audio device unavailable” or “Session startup failed.”Speaking about the reliance on technology, yes, some creatures have to rely on technology, like wheelchairs, pacemakers, etc. as well to be able to live.
-
What does AI even do other than produce slop and make people reliant on technology
It allows a blind creature, who has to operate a Linux installation without working speech synthesis (because the maintainers messed that up and didn't care about accessibility), even attempt to debug the mess in the first place and have an operational, accessible machine:
Want to debug it?
You can’t—because you can’t hear anything.
So you grab your phone, take a picture of the screen, feed it to an image captioning AI, and hope it tells you whether the error dialog says “Audio device unavailable” or “Session startup failed.”Speaking about the reliance on technology, yes, some creatures have to rely on technology, like wheelchairs, pacemakers, etc. as well to be able to live.
@aronowski @nicuveo That's fair. If you need AI for something like that I won't judge. Especially when it is just extracting information.
When I said "AI" I meant more of Generative AI like Claude or ChatGPT.
If someone needs technology for something that's fine too, but some people are losing what they have because they use tech they don't need. Like if a person who 100% can walk uses a wheelchair constantly. Especially when it comes to things like Gen AI.
-
@nicuveo Richard Stallman just did this toot opposing breathalyzers in cars: https://mastodon.xyz/@rms/116581752793267275
It's all tied together.
@indigoparadox @nicuveo I think the kill switch he’s opposing is not a breathalyzer but a driver-facing camera that’s supposed to detect “impairment” using some bullshit algorithm. It’s legit creepy and invasive - some cars already have them to beep at you if you stop watching the road while the assisted cruise thing is on. But there’s a bill proposing making them mandatory for all new US cars
-
@nicuveo A lot of analyses of drunk driving focus on the deaths and injuries, but they ignore the hidden costs of enforcing driving regulations. In fact the strict regulations create a dependency culture that encourages more drunk driving. /sarcasm
@SecondUniverse @nicuveo Jesus, you are TOO good at that. Please promise to only ever use your powers for good.
-
i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs. fuck that! ethical concerns front and fucking center! it is very revealing that tech is currently in such a state that the quiet part can be said out loud without any pushback.
@nicuveo "aside from all the murdering, Ted Bundy was a pretty good guy!"
-
@nicuveo@tech.lgbt I'll gladly stop using LLMs (professionally and personally) provided someone does the same tedious work they do for me at the same cost (i.e. near zero).
Ethics gotta step aside when 'I have to eat' and 'I can't think straight' is the cause. -
every conversation about the potential usefulness of AI, divorced from ethical concerns, is just this dril tweet
-
i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs. fuck that! ethical concerns front and fucking center! it is very revealing that tech is currently in such a state that the quiet part can be said out loud without any pushback.
@nicuveo My "ethical concerns aside" has been "It's still shit"
-
@nicuveo@tech.lgbt I'll gladly stop using LLMs (professionally and personally) provided someone does the same tedious work they do for me at the same cost (i.e. near zero).
Ethics gotta step aside when 'I have to eat' and 'I can't think straight' is the cause. -
i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs. fuck that! ethical concerns front and fucking center! it is very revealing that tech is currently in such a state that the quiet part can be said out loud without any pushback.
@nicuveo to be fair there was an incredible amount of pushback which was heartening to see
-
@pgcd@mastodon.online @nicuveo@tech.lgbt Surprisingly, slaves need to be sheltered, fed, clothed. My GPUs only need a bit of space in the PC case.
So no. I couldn't afford a slave, even if I were that evil.
Hence, argument invalid. -
What does AI even do other than produce slop and make people reliant on technology
It allows a blind creature, who has to operate a Linux installation without working speech synthesis (because the maintainers messed that up and didn't care about accessibility), even attempt to debug the mess in the first place and have an operational, accessible machine:
Want to debug it?
You can’t—because you can’t hear anything.
So you grab your phone, take a picture of the screen, feed it to an image captioning AI, and hope it tells you whether the error dialog says “Audio device unavailable” or “Session startup failed.”Speaking about the reliance on technology, yes, some creatures have to rely on technology, like wheelchairs, pacemakers, etc. as well to be able to live.
@aronowski OCR existed before "AI"......
-
i am so tired of "ethical concerns aside" being a phrase i see every single time someone tries to defend the use of LLMs. fuck that! ethical concerns front and fucking center! it is very revealing that tech is currently in such a state that the quiet part can be said out loud without any pushback.
@nicuveo this bizzare pattern of people openly undermining their argument then acting like they did not just undermine their argument
-
every conversation about the potential usefulness of AI, divorced from ethical concerns, is just this dril tweet
@nicuveo @joshsutphin Needless to say, I agree.
Cassandra is only carbon now (@xgranade@wandering.shop)
Content warning: alc, dui
The Wandering Shop (wandering.shop)
-
Just catching up on these messages now. I read some stuff between the lines in your original post, and came in guns blazing, but that's not what this situation needed. I'm sorry for what you're going through. I'm unemployed right now myself. I worry almost every day that my partner will lose her job before I get the surgery that'll let me work again. We're not so different.
I said a lot of "planet health" reasons LLMs suck, but I think the important thing in this context was actually the last one: collective bargaining power. This world *does* have enough resources to go around, but the people at the top with the power to hoard don't have a willingness to share. They are why the rest of us get a musical chairs, you-or-me economy, fighting each other over the scraps.
LLMs exist not to fix the understaffing problem, but to create and enable it. But it's a fair question where that leaves you as an individual actor, and I want to give you an honest and useful answer.
I was unemployed for nearly a year
I think I got dinged in a couple of interviews for not loving the bot
I applied to ethical places and didn't make the cut
I got a Claude subscription and learned to be literate in LLM code generation
The job I eventually got was one that also wanted that
Never going to hate anyone for using this stuff to stay in work, I just hate that it's a requirement
Hoping for better days for both of you
-
@aronowski @nicuveo That's fair. If you need AI for something like that I won't judge. Especially when it is just extracting information.
When I said "AI" I meant more of Generative AI like Claude or ChatGPT.
If someone needs technology for something that's fine too, but some people are losing what they have because they use tech they don't need. Like if a person who 100% can walk uses a wheelchair constantly. Especially when it comes to things like Gen AI.
@VANTABlack2000 @nicuveo I've already heard of cases where employees were pressured to use LLMs as part if employee performance assessment. I feel for them, if the case is that the tool makes the work harder than easier, because not only one should be able to use the tools they want and tailor them to their needs, but the models might also produce garbage. I've seen a case where an AI agent was analyzing a codebase, which one buddy knows by heart, and exhausted all the token credits, trying to do anything. Would be better if everyone was using the tools that work best for them.
That said, what do you think of the cases where an agent can synthesize software, that a non-programmer can show as a proof-of-concept, whereas before AI such person wouldn't be able to write any code at all? Think of creatures like me, who get frustrated when trying to learn to program, who enjoy working with paper documents and would rather use Excel for data crunching than (unsuccessfully) write Python scripts?
