software-related AI Video has arrived - It looks incredible

The new chips coming out this summer are 25-30x more efficient, generating an image isn't so bad... but AI gets a lot more complicated than an image. When you want it to do stuff like cure cancer by examing all 3 billion of your specific dna pairs, or automatically creating a piece of software or video game, etc

In real life those games take hundreds of people like 8 years to make, so even if AI can do it quickly, yeah its using a lot of brain power.

AI Uses so much electricity these companies want to build their own nuclear power plants just to run it.

www.theverge.com

Microsoft is going nuclear to power its AI ambitions

Microsoft is hiring someone to lead its nuclear strategy.
www.theverge.com www.theverge.com

www.theregister.com

AI turning nuclear power plants into prime DC real-estate

Cheap low-carbon energy? What's not to love...
www.theregister.com www.theregister.com
 
Jack Lynt said:
Which version of ChatGPT did you use?
Click to expand...
Let me check. It was the first free one: 3.5. I'd be curious if someone with an upgraded version got the same answers, about Bach, prison, and the Well-Tempered Clavier (book 1). I think what might have confused it is that, although it is a cool and common story, that the WTC was composed in prison, there is only one bit of scant evidence for It. Here is Christoph Wolff, from Johan Sebastian Bach: The Learned Musician;

. . . a useful hint is provided by Ernst Ludwig Gerber (whose father, Heinrich Nicolaus Gerber, studied with Bach in Leipzig during the 1720s) when he relates that Bach wrote his Well-Tempered Clavier, Part I, “in a place where ennui, boredom, and the absence of any kind of musical instrument forced him to resort to this pastime.”74 Though we cannot take this to mean that the work was begun and completed during Bach’s imprisonment, a substantial portion of the twenty-four preludes and fugues may well have originated in this unhappy venue.
Click to expand...
 
Last edited:
Spike said:
Let me check. It was the first free one: 3.5. I'd be curious if someone with an upgraded version got the same answers, about Bach, prison, and the Well-Tempered Clavier (book 1). I think what might have confused it is that, although it is a cool and common story, that the WTC was composed in prison, there is only one bit of scant evidence for It. Here is Christoph Wolff, from Johan Sebastian Bach: The Learned Musician;
Click to expand...

Claude is free and better than GPT 3.5, but if its serious research, then none of them are "reliable"
I've tried using GPT 4 for learning history and then end up looking dumb quoting wrong facts.

claude.ai

Claude

Talk with Claude, an AI assistant from Anthropic
claude.ai claude.ai
 
Last edited:
Voice replies without a delay now!
Sounds human enough to fool some people. woah.

https://www.reddit.com/r/nextfuckinglevel/comments/1cradtv/open_ais_gpt4o_having_a_conversation_with_audio

https://www.reddit.com/r/singularity/comments/1cr58kz/people_trying_to_act_like_this_isnt_something


That's how the actors are gonna sounds in Sora with current gen tech.

Keanu Reeves GIF


"Shortcomings" are temporary. Took about 6 weeks this time for a big announcement, it was actually quiet for a bit.
 
Last edited:
Last edited:
sfoster said:
Wow hard disagree, Its astounding someone could listen to that and think its a 'routine update'
Click to expand...
But it is just "routine". It's the same pre-packaged samples they keep throwing at us, completely devoid of any real-world context. While there may be some humans who talk like that in real life, to me it sounds 100% computer generated - nothing more than the audio version of the utterly useless "AI" customer service bots that can never give you a meaningful answer to a real problem. That first clip's whole conversation is one cliché statement after another, interspersed with canned laughter, so yes, it looks (and sounds) like an utterly banal "routine" update to me, too. :no:
 
Tough crowd.
indietalk said:
Mark me down for just another Monday
Click to expand...
If you were blind or knew anyone blind I seriously doubt you'd feel that way.
The brand new ability to pull out your phone and have it describe in real time what's happening is LIFE CHANGING


Edit to add: Soaking in these responses, I'm actually even more impressed and blown away.
The fact that we live in such a time that things like this could be regarded as routine is absolutely WILD.

What a time to be alive.
 
Last edited:
Now that it has real time chat capablities, it functions as a free universal translator:


To me this is NOT just another monday!
Universal translators are a big deal, sci-fi come true, but thats just me.

The technology even exists to translate and have the ai sound like YOUR VOICE in the translation, how wild is that.
I believe they're a bit wary of using that part though since it's so easy to abuse.

Or you can use the camera to read out loud and translate books/text/webpages in other languages and read those to you.
Maybe you go to a restraurant and the menu is in french, point it at the menu and listen to the translation.
 
Last edited:
sfoster said:
The fact that we live in such a time that things like this could be regarded as routine is absolutely WILD.
Click to expand...

Meh ... :hmm:

For me, it's "routine" because there's nothing new about it, nothing special - we (humans) have been doing that for centuries. But we've also being doing it in a way that adds value to the interaction, not simply regurgitating stock phrases.

But again, these are all marketing videos - where are the examples of real people in real situations using the technology? I want to see it cope with the kind of carry-on that got the Dublin-New York portal shut down this week. :D
 
Last edited:
CelticRambler said:
But again, these are all marketing videos - where are the examples of real people in real situations using the technology? I want to see it cope with the kind of carry-on that got the Dublin-New York portal shut down this week. :D
Click to expand...
It's not gonna 'cope' with that. If you flash ChatGPT your dick it'll *BAN YOU* for violating the terms of service and being a degenerate.
I'm a paid subscriber, I expect to have access to ChatGPT Omni within the next 2 weeks.
 
sfoster said:
The brand new ability to pull out your phone and have it describe in real time what's happening is LIFE CHANGING
Click to expand...
Don't we already have enough of people pulling out their phone to "capture the moment" and missing out entirely on the "moment" happening right in front of them because they're focused on a screen instead? And then millions of petabytes of crappy videos posted online not actually showing the moment something happened, but the minutes after whatever "it" was. This all comes across as a desperate attempt to find use cases for an idea where they're not particularly needed. If they were, chances are we'd already have come up with an effective solution ... and in many situations we have. I doubt any blind person's life will be changed by listening to a computer-generated voice telling them how great it is to be waving a phone at a flag on a pole, or to be told that ducks are doing duck things. :rolleyes:
 
CelticRambler said:
Don't we already have enough of people pulling out their phone to "capture the moment" and missing out entirely on the "moment" happening right in front of them because they're focused on a screen instead? And then millions of petabytes of crappy videos posted online not actually showing the moment something happened, but the minutes after whatever "it" was. This all comes across as a desperate attempt to find use cases for an idea where they're not particularly needed. If they were, chances are we'd already have come up with an effective solution ... and in many situations we have. I doubt any blind person's life will be changed by listening to a computer-generated voice telling them how great it is to be waving a phone at a flag on a pole, or to be told that ducks are doing duck things. :rolleyes:
Click to expand...
For real? Okay how about this

Blind person walks to the grocery store, wants to know which milk is the 2%.
They can pull out their phone and it will guide them to the right container.

Yes there are other solutions, you can video chat with a human, you can ask a person, but INDEPENDENCE gained from technology is priceless.
That feeling of constantly depending on other people sucks, people want to feel independent and technology enables that feeling.
 
sfoster said:
It's not gonna 'cope' with that. If you flash ChatGPT your dick it'll *BAN YOU* for violating the terms of service and being a degenerate.
Click to expand...
So there you are: between the innate limitations of the software and the terms-and-conditions imposed by the cultural norms of some random developers, it's restricted to recycling banal information and displaying phoney emotions. Hardly groundbreaking stuff.
 
CelticRambler said:
So there you are: between the innate limitations of the software and the terms-and-conditions imposed by the cultural norms of some random developers, it's restricted to recycling banal information and displaying phoney emotions. Hardly groundbreaking stuff.
Click to expand...

I believe we'll see radical, unprecedented change in our lives throughout the 2030s

You know I'm a big believer in this stuff, but even *I* would not be surprised if the next 50-100 years or more continue to be limited to "phoney emotions"

What you're asking for there, for "genuine emotions" is such a tall order it's not even in the scope of anyones ambition.
AI scientists don't care about creating life that feels emotions, that's entirely BESIDE THE POINT.

AI is about problem solving, you give it any problem (e.g. 'draw me a picture of a pitbull in a suit') it gives you a solution.
The idea of actually creating life and it having feelings is something for another century perhaps.

In short nobody cares what it feels; they only care what it can accomplish

CelticRambler said:
Let's see the video, featuring a real blind person in a real store ....
Click to expand...

I'll still got blindman props, the cane and everything.
I'll give it a shot once I have Omni videochat access.
 
Last edited:
Every time I have to deal with an AI chat bot, I end up wanting to tear out my hair and never being able to get the help I need.

Same is true with trying to report abusive people on FB, X/Twitter, etc. - all I get is a canned response that does nothing to resolve the issue but allows the platform to say it's been "handled." Same is true about banking or credit card AI bots, plane reservations, etc. Not one time have I been able to resolve the issue that I called about without hanging on for god knows how long to get a real person, even though I'd love to get done in 5 minutes by talking to the AI bot. It never actually works, except that I'm sure they get plenty of callers to simply hang up in frustration - which save employees' time, right?
 
You must log in or register to reply here.
Back
Top