I hope doctors are the first to lose their jobs btw, or rather, i hope it will flush out the garbage / fake ones, because there are so many of them and they add nothing in comparison to just googling or asking an AI. Of course, every profession has this issue, but not every profession handles people's lives and makes a really good living of it while practically doing nothing for the person. If someone's earned a really great reputation as a doctor he will never be out of a job and we will always need them, they don't need to worry.
I've seen a video on Twitter (can't find it right now) where a doctor is asked for a diagnosis and he basically goes through a decision tree from memory, it's not much different than what AI is doing.
But doctors are not going to lose their jobs anytime soon, they're safe for two reasons:
- Demand for human to human contact will stay high if it concerns personal health/illness etc. which is a psychological thing
- The profession is one where human+AI cooperation is most possible, so the best and most demanded doctors will be those with a personal AI assistant to augment their capabilities. A super-diagnosing Human-AI duo - so if you are a doctor best get on that train early.
People will adapt to AI really quickly. Anything that simplifies our lives we will be more likely to choose, especially if it's something that's also: super instant, free, doesn't involve human interaction. I agree that we still likely prefer to hear it from a real human, but it's only because talking to an AI is completely new to us, it will absolutely change, and quickly because we will use it for a lot of things so it won't be awkward for very long.
Anything bank related was once upon a time also human-human, if you go to bank nowadays you will rarely see young people there, it's only really old people that were too old to adapt to using computers or mobile phones for bank businesses. I'd say most people prefer it its not human-human interaction when it comes to dealing with money/bank issues, but it absolutely wasn't always the case, we adapted and computers or our phones gained our trust.
I've seen a video on Twitter (can't find it right now) where a doctor is asked for a diagnosis and he basically goes through a decision tree from memory, it's not much different than what AI is doing.
But doctors are not going to lose their jobs anytime soon, they're safe for two reasons:
- Demand for human to human contact will stay high if it concerns personal health/illness etc. which is a psychological thing
- The profession is one where human+AI cooperation is most possible, so the best and most demanded doctors will be those with a personal AI assistant to augment their capabilities. A super-diagnosing Human-AI duo, so if you are a doctor best get on that train early.
This. Medical AI tools will be there soon and probably be to good not to use.
I will not be out of a job soon. But likely specialised nurses will take over, they are a lot cheaper.
You work under assumption that doctors, unlike every other profession in the world knows everything always they ever learned. Imo huge mistake.
No I work under the assumption he will consult on that which he doesnt know off hand, defer to specialist on that which is he is not versed, and will combine past info with current info and work out a new 'emergent property' of a diagnosis personalized for me.
Its not only the knowledge from books I look for in the a doctor, its knowledge that cant be learned from books: personalized for me weighting all the combined hint and clues of the bigger picture of my unique personal makeup and issues, based off the knowledge in books combined with the wisdom of practice and experince. That is what I look for.
as long as it's not trained on her voice I personally don't see a problem
Do we sue now every time some actor declines a role and you get someone similar looking?
what about movie tie-in games where actors decline to voice and you get a soundalike?
and if you're a voice actress with a voice too similar to her's I guess you're fucked?
Likot Mosuskekim, Woodcutter cancels Sleep: Interrupted by Elephant.
@Surray That was my first reaction too, but something is fishy. She requested info on how they made it sound so similar, and their reaction was to take it down instead of saying “well, we used actress so and so, fuck off”. So something fishy might have been afoot.
They would have sampled her voice and generating a text to speech model without permission, which would shit storm. Sam Altman and OpenAI seem like a real asshole company I guess the prerequisite for successful big tech.
This seems pretty crazy to me, interaction with degenerates, AI generated responses in realtime, Voice cloning in real-time, animation lip sync (shitty but still not bad).
I just got into a discussion with a friend about how AI will replace coders/programmers.
As he is not one, he thinks my job consists of: have clear and detailed programming goal, program to make that happen, profit.
While I think AI will one day change how I work, it wont replace me at least not all of my job, just the tedious coding part. Why? Only the very end of the road in any contract/environment is doing the actual coding.
90% of my job is talking to customers that dont know how to articulate what they want, to figure out what they MEAN they want: I made it do what you explicitly told me to make it do in the way you want. Do you want me to change it to what you want now that you see doing it that way does not give the desired results you thought it would? Or leave it as what you told me to make it as, as they are not the same.
The coding is just the last goalpost after all the 'humans trying to convey humans ideas' part is done.
Example:
I might become a Prompt Engineer vs a Application Engineer. As I am still needed to know what to tell the AI to make, as they are not knowledge enough of it to tell it what to make well.
Them: Make the GUI code show the PDF pages bigger. They are too small.
Me: I know this means make the NWX birdeye view specifically, show the thumbnails of the pages from the 'actuals' folder that is a database reference to the PDF that is not the PDF itself, that pulls up the full_[idnumber].png of dynamically available PDF png renders (not the PDF itself, they need ripped CMYK preview. What they send and what it prints as ripped can be different), in a popup window to inspect the pages, so they can approve or reject them. Needs me using context of what they use it for and we talked about in the past they still dont 'get' but I know what they mean, and what they probably mean.
They dont even know it's not the PDF, but the print render on the back end. or that it needs the full.png and not the preview.jpg to tell the AI what they need it to do. Just what they think its doing and would tell it the wrong thing to do, and it does exactly that wrong thing they specially told it: make the mono.jpg (they think as a PDF, which is wrong) I see bigger (which is wrong, it needs the full.png not PDF OR jpg).
--
Good luck AI, as humans have been confusing/badly conveying to this human for 17 years (my time here at this job) on what they really want, vs what they say or tell me to do, and my job is mostly figuring out the difference from knowing that customers way of speaking specifically. Wish you all the luck in the world understanding humans, that others can barely understand when telling you ideas of 'what they want'.
Will it get there? Maybe, but humans need to be better at saying what they want, for it to know what they want.
Last edited by |DXWarlock on Thu, 23rd May 2024 02:54; edited 1 time in total
@friketje
How is that bad? A real time contextual tutorial for a game?
Of all the things AI could be feared/mocked/jested about trying to do. This seems the more innocuous good fit for it.
When I started playing back in eeeearly alpha (back when it was just barely publicly out. No hype, or tutorials, or videos on how to play yet), I would have loved that. You know how long it took me to figure out how to use crafting, and I just punched blocks and trees for like an hour? With no idea how the 3x3 grid worked?
Yeah of course they stole the voice. Remember the interview where the girl boss CTO or whatever from OpenAI basically admits they train on copyrighted videos?
I mean if you put up a recording of Sky next to a recording of scarlett johansson you can tell it's not her, quite easily
It's just very similar, though some might say Sky sounds more like Rashida Jones
I'd say it's a bit of a mix, which if it was a real person voicing her, might be? Plenty of people on the planet that sound similar
Fair enough though to get to the bottom of it. OpenAI might just be mixing together different celebrities secretly to get to their results which would be fucked up
Just think a home user can generate a cloned voice of Biden / Trump on a whim downloading some opensource stuff, requires barely any hardware screwing around. Wonder what OpenAI can and would do?
What it can or cant do isnt the problem. Its people IMHO.
Generate all the fake Trump/Biden voice or even video of them all you want and force it into my internet feeds. if I can't confirm or verify/cross reference the source or news of it. I dont believe it.
The failure isn't how good AI can trick us, its how good we are at being tricked. Too many people will see a single source of a thing, believe it, and tell OTHER people and they believe it too, without any of them looking into/looking it up for even a single reputable corroborating source.
Dont need AI for that. Vague articles with no merit and just text causes that now. Sure it makes the people actively listening for a pipe song from the piper more ready to join the rat conga line. But if it wasnt AI it would be some other thing they hear the melody of with the same result.
Apparently Hurrdurrvarr Derpderrson is a “folk hero”, fighting the next frontier of the wokeh line in the sand.
This shit came today, after it has already been clear its frivolous nonsense, but apparently “Across X and news reports”, “Lawyers say she might have a case”
TL:DW google's amaZOMG ai overview sometimes takes into account sarcastic/trolling shit and presents it along with valid suggestions.
As in, add glue to your cheese sandwhich so the cheese doesn't slide, use gasoline for a spicy spaghetti, etc
“The examples we’ve seen are generally very uncommon queries, and aren’t representative of most people’s experience using Search,” a Google spokesperson said in a statement.
and the fact it isnt just some app, but a complete retarded 10 usd made in china cheap plastic device they charged 200 usd for ... says everything u need to know about the intelligence level of backers and investors ... total retards
there's so much fake shit when it comes to AI right now, just check any app store (android, apple).. i don't understand how it's even allowed, like companies claiming they're OpenAI when they're just scamming people.
Retardoogle is manually removing embarrassing hallucinations of its “A”“I”, such as “Obama is the first muslam USA president”.
Remember when companies used to sell “computer vision” products that “knew” how to solve captchas, but it was actually indians or chinese doing it in secret? Google is about to announce a new industry and become a titan in it…
I wonder if some retardican (since that's the land of litigation) will try and go ahead using/doing one of those fucked up results to then try and get some millions from google in court
boundle (thoughts on cracking AITD) wrote:
i guess thouth if without a legit key the installation was rolling back we are all fucking then
It’s a catastrophe waiting to happen. Apparently, this shit by Retardoogle is recommending people put glue in their home-made pizzas. Imagine some kid reads...
Onion, joke subreddits, etc. It seems to have a talent for finding the most obscure joke articles and bringing them into prominence. And Pundurr Suchai is making it like this is the best thing ever
Signature/Avatar nuking: none (can be changed in your profile)
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
