Sea Speed said:

---

I want to know how to make money off of it. Where do I invest?

---

Ask ChatGPT

Adverse Event said:

---

"This is the worst AI will EVER be"
Truer words neer spoken

This is the worse AI will ever be.

The Music industry is completely gonna be turned upside down.

Every artist will have their own custom trained model, and Spotify will own this space why?

🧵 A thread

pic.twitter.com/3Jqz32JjkP

— Linus (●ᴗ●) (@LinusEkenstam) March 26, 2023

Only thing I don't agree with here is prognosticators staying "in a few years, x,y,z..." AI advances happen exponentially. Artists will have trained AI's in the next few weeks or months, or their producers will.

Maybe sooner. I thought this stuff would be happening fast...but it's almost to the point that EVERYONE needs to quit what they're doing and train themselves and their AI to be a productive member of society... maybe this is just FOMO, but it doesn't feel like it.

---

This is awesome.

Artists will be paid for their voice, then it will be used by creators possibly to whatever they want.

Who knows what future businesses look like. Humans still like to see a face and a human behind a production, so that won't go away.

The group Gorillaz is a totally animated pop band. It has no people, just some music producers and cartoons. We'll see more of that with AI.

And talented musicians will make whole albums solo.

Really cool tech.

I look forward to taking songs with ****ty vocalists and adding better vocals.

Sea Speed said:

---

I want to know how to make money off of it. Where do I invest?

---

Global X (BOTZ) traded on Nasdaq. About $25 a share. Up 33% in the last 6 months.

You could have super ugly people with amazing voices bebthr voice behind some hot person with stage presence and charisma. The public facing person won't actually need any talent.

If you’re having trouble keeping up with Video AI😅, there have been 5 state-of-the-art generative video models released *in last 7 days*: 🤯😎🧵 pic.twitter.com/cki9SMfozr

— Philipp Tsipman (@ptsi) March 26, 2023

Text to video is here now as well. It's about to become trivial to make movies.

I worry about the consequences for educators.

Just out of curiosity, I asked it to write my Master's thesis, giving the premise and a summary of my primary research and requested it give 30 peer-reviewed citations.

It did it instantly.

It took me two years to compile and present that research. That was twenty years ago, so things are easier to find now anyways, but the threat here is that it will hijack research and discovery.

Even if students don't straight up plagiarize the AI outputs, they can certainly just make the AI work the backbone of their manuscript - just move some stuff around, add some content, and boom - they can complete an entire thesis in a single afternoon.

I think there are going to be massive benefits to this, but there are going to be some costs and it's going to take someone a lot smarter than me to figure out how to manage those consequences.

Nanomachines son said:

---

Adverse Event said:

---

"This is the worst AI will EVER be"
Truer words neer spoken

This is the worse AI will ever be.

The Music industry is completely gonna be turned upside down.

Every artist will have their own custom trained model, and Spotify will own this space why?

🧵 A thread

pic.twitter.com/3Jqz32JjkP

— Linus (●ᴗ●) (@LinusEkenstam) March 26, 2023

Only thing I don't agree with here is prognosticators staying "in a few years, x,y,z..." AI advances happen exponentially. Artists will have trained AI's in the next few weeks or months, or their producers will.

Maybe sooner. I thought this stuff would be happening fast...but it's almost to the point that EVERYONE needs to quit what they're doing and train themselves and their AI to be a productive member of society... maybe this is just FOMO, but it doesn't feel like it.

---

The entirety of the music and movie industries are about to be blown apart. People will be able to create music and movies based upon prompts from AI trained on specific musicians, actors, and more. Voice acting is 100% dead now and we're about 2 years away from AI making motion video with ease.

It's absolutely not fomo, it's the biggest change since the invention of computers themselves. The rapid advancement is going to crush large bureaucracies that are unwilling to change fast. I've been talking with everyone at my office to get this moving forward, but the resistance is high.

---

There's only value of you can own the rights to said music and movie.


Unless society decides to to allow ai vocals, music , art etc.to be copyrighted, it's not all that valuable. A studio isn't going to sink money into a project unless they can be certain they can control it and make money off of it.

People are going to learn less so when civilization resets, great swathes of human accomplishment and knowledge will be gone.

It's moving so fast.

Wow.

Will have to sit and think.

How can I leverage it besides just generating media content for consumers.

This. What can I do as a non tech person to use this to my advantage.

Sea Speed said:

---

People are going to learn less so when civilization resets, great swathes of human accomplishment and knowledge will be gone.

---

That's awfully pessimistic.

I don't see how AI eliminates previous knowledge. It could be reasonably argued that it preserves that knowledge. The lifeblood of AI is historical data. It just synthesizes what is provided, so it's a kind of organized regurgitation, for lack of a better phrase.

An optimistic view of this technology would be that it can be a tool for innovation rather than a lazy crutch that limits innovation.

And although this en-vogue belief that civilization is headed for some sort of reset has some merit because there's a lot of scary **** in play right now, the data is safe. Outside of nuclear annihilation or some other human extinction event, I don't see a scenario where we just lose track of all scientific advancements.

Even if there's a large-scale global war again, it doesn't throw us back to the stone age. In the past century, World Wars have actually spurred technological development. It's a ****ty and brutal way to advance our technology, but it was effective in doing so the last two times. Hell, for us here in the US, those wars created our wealth and global power.

If literally everything can be done with a computer program, there is no need for people to learn the things the computer does, only for them to learn how to get the computer to do the thing they want it to do. Sure some people will know how to do the things, but I see this leading to fewer of them.

Sea Speed said:

---

If literally everything can be done with a computer program, there is no need for people to learn the things the computer does, only for them to learn how to get the computer to do the thing they want it to do. Sure some people will know how to do the things, but I see this leading to fewer of them.

---

We are in agreement with the exception of one point. Not literally everything can be done with a computer program. Even the most robust, insanely well-written AI code is limited by the data it draws from. and that data is finite.

New discoveries borrow heavily from previous ones, but without human ingenuity, nothing is added to our collective knowledge, and innovation stops.

Imagine if Einstein had an AI program in 1920 or whenever and wanted to run his ideas regarding general relativity past it, and he asked Chat GPT about fourth dimensions and curves in time and space and the speed of light, etc. That version of AI would only be informed by Newton's gravitational laws and would have no idea what he was even asking.

Computers are tools that humans made. Human ingenuity can do things that they cannot without us informing them of what our brains figured out.

That's why it's artificial.

My hope is that AI makes us ambitious. My fear is that it makes us lazy.

You're right, I should not have used the figurative definition of literally.

It's gonna be a matter of time before they combine the chat gpt chatbot with this voice cloning technology.

You'll have server farms of chatbots making automated phone calls, taking to people, doing sales, customer service, checking in on elderly people. Checking in on patients. Getting responses to whatever the bot is programmed to ask.

I think we'll see this voice cloning be able to emulate regional dialects. Make you feel like you are talking to someone with your native dialect.

You might have a giant databank of voice profiles. So the algorithm selects a voice that it thinks you would be most responsive to based on your profile. Whatever that may be.

Indian call centers might get buggy whipped. Sales calls, scheduling calls, checking in on people over the phone. All that could potentially be automated.

Computer would do it perfect every single time and there would be a record of every customer interaction. Humans only get involved if there's an issue.

We'll get to a point where we won't be able to tell whether we're talking to a bot or a human.

Listen, all I want to do before I die is become Dr. Lexus



It's gonna be like three weeks from now at this rate.

Do you think AI will ever be developed that can have ingenuity or eureka moments? Or is that ability biological?

Sea Speed said:

---

You're right, I should not have used the figurative definition of literally.

---

Haha. No worries, man.

I find this stuff really fascinating.

I play around in Chat GPT a lot and it's mind-blowing to me. Sometimes it is so spot on it's kind of scary and I learn something, and other times it's clunky and filled with inaccurate, useless data. So, I guess it's a bit like F16

I think it can connect dots already. But it has to have the dots to connect.

TxAgswin said:

---

I think it can connect dots already. But it has to have the dots to connect.

---

Is connecting the dots the same as a eureka moment? I don't know...

All of those technologies exist already.

Somebody just has to pull them together and it wouldn't be that hard.

It's actually a really good idea.

Stat Monitor Repairman said:

---

It's gonna be a matter of time before they combine the chat gpt chatbot with this voice cloning technology.

You'll have server farms of chatbots making automated phone calls, taking to people, doing sales, customer service, checking in on elderly people. Checking in on patients. Getting responses to whatever the bot is programmed to ask.

I think we'll see this voice cloning be able to emulate regional dialects. Make you feel like you are talking to someone with your native dialect.

You might have a giant databank of voice profiles. So the algorithm selects a voice that it thinks you would be most responsive to based on your profile. Whatever that may be.

Indian call centers might get buggy whipped. Sales calls, scheduling calls, checking in on people over the phone. All that could potentially be automated.

Computer would do it perfect every single time and there would be a record of every customer interaction. Humans only get involved if there's an issue.

We'll get to a point where we won't be able to tell whether we're talking to a bot or a human.

---

the real voight-kampff test will be to call it a benchode or untouchable Dalit and see what happens.

The Green Dragon said:

---

TxAgswin said:

---

I think it can connect dots already. But it has to have the dots to connect.

---

Is connecting the dots the same as a eureka moment? I don't know...

---

I guess sometimes it is. The scene in the Steve Jobs movie (the Ashton Kutcher one) comes to mind - where he kind of "sees" the home computer is a connecting-the-dots kind of eureka moment where the dots existed already.

However, Einstein's moment of redefining mankind's understanding of the universe was probably a eureka moment where there were no dots outside of that guy's brain.

Sometimes humans have to create the dots.

David Happymountain said:

---

Stat Monitor Repairman said:

---

It's gonna be a matter of time before they combine the chat gpt chatbot with this voice cloning technology.

You'll have server farms of chatbots making automated phone calls, taking to people, doing sales, customer service, checking in on elderly people. Checking in on patients. Getting responses to whatever the bot is programmed to ask.

I think we'll see this voice cloning be able to emulate regional dialects. Make you feel like you are talking to someone with your native dialect.

You might have a giant databank of voice profiles. So the algorithm selects a voice that it thinks you would be most responsive to based on your profile. Whatever that may be.

Indian call centers might get buggy whipped. Sales calls, scheduling calls, checking in on people over the phone. All that could potentially be automated.

Computer would do it perfect every single time and there would be a record of every customer interaction. Humans only get involved if there's an issue.

We'll get to a point where we won't be able to tell whether we're talking to a bot or a human.

---

the real voight-kampff test will be to call it a benchode or untouchable Dalit and see what happens.

---

In a couple of years we'll be on here talking about our preferred method of outing a bot.

There'll probably little tweaks you can say, or little hacks that will cause the machine to **** up and give itself away. It'll be the reverse of Wozniaks' tone device that hacked the phones. It'll be the reverse of that. Trying to determine whether something is man or machine.

The developers will be constantly tweaking it to patch it up. Make it more foolproof. Harder to detect. They'll be some controversy where people will demand that companies disclose whether a caller is automated or not. By the time the issue makes it through the judicial / regulatory system, it will be too late. Government and every industry will be using the technology and nobody will protest that much. It will be just the way it is.

I don't think we are too far away from an app that will FaceTime somebody with a flawless digital avatar and say anything the programmer wants it to say.

It's gonna get wild out there.

Think how bad scamming will be with AI voice cloning. We arent going to be able to trust anyone over the phone. You're going to have to have some sort of 2 step app authentication just to trust your speaking to your friends and family. That will morph in to needing some sort of bio authentication for everything to prove you're a human, and beyond that, the human you're claiming to be. Tbis whole thing is going to get real dark real fast.

We will not trust anything we don't see with our own eyes. I mean nothing. You won't be able to trust a phone call, video chat, news story, press conference etc etc.

Next pandemic we'll have humanoid chatbots calling our phones every day to check our compliance with public health measures recommended by the CDC.

If you don't answer for the chat bot they'll send someone out to do a wellness check.

You know ... it's the best way protect pubic health and track outbreaks.

People that don't take the vaccine are upped to two calls a day.

If you don't like it they'll shut off your internet until you complete a wellness check, get stabilized and get in compliance with CDC guidelines, or perhaps your employers' guidelines.

It's one thing to answer an email to check in on an app. But the ability to have to call and converse with a machine and have the computer assess whether what you are doing is authorized and same.

Imagine the driver of a commercial vehicle having to converse with a video chatbot that would assess whether they were tired, drunk, on dope, on pills, high, hungover or whatever the case. the machine would determine whether you were fit to drive.

Could be used for anything really. There may be a time where you show up to work and have an automated competency test that would determine whether you could work that day. OSHA regulations require that anyone working at this facility pass an automated competency screening.

I can see insurance companies requiring this in certain industries.

Sea Speed said:

---

Think how bad scamming will be with AI voice cloning. We arent going to be able to trust anyone over the phone. You're going to have to have some sort of 2 step app authentication just to trust your speaking to your friends and family. That will morph in to needing some sort of bio authentication for everything to prove you're a human, and beyond that, the human you're claiming to be. Tbis whole thing is going to get real dark real fast.

---

When is the last time you answered a call from a number you didn't recognize?

If somebody that sounds exactly like your brother-in-law calls you up and asks you for your banking information, don't give it to him.

It is healthy to worry about security threats presented by new technology, but damn.

Quote:

---

Imagine the driver of a commercial vehicle having to converse with a video chatbot that would assess whether they were tired, drunk, on dope, on pills, high, hungover or whatever the case. the machine would determine whether you were fit to drive.

---

If the technology was accurate, that would be great. It would save lives!

You cannot be serious.

I don't know, I think folks are giving this tech way too much credit.

The entertainment industry isn't going to die. People want to hear real music performed by real humans. It's why guitar players like tube amps vs digital. The differences will be very noticeable.

Call centers, yeah I can see that but, it will be pretty obvious you are talking to a bot.

Some people think self driving vehicles are just around the corner but, after watching a video showing off Tesla's latest software, even that is still a very long way away.

A computer isn't going to fool anyone with half a brain, its just not.

You realize number spoofing is already a thing right? And I'm not worried about me, but millions are stolen from people every year over the phone. If you can make the caller sound like someone that isn't from a third world and can have a real time conversation with a mark, people will be tricked and taken advantage of even more than they are now over the phone.

Classroom teachers are at risk of being replaced by video chatbots tailored to the individual student.

You might have a kid that has the same math teacher all the way through every math subject. Same teacher so there's consistency in learning. The chatbot would know exactly what the kid knows and didn't know in real time. there would be no way to half ass it or just get by. You d be required to demonstrate and apply knowledge. You'll have people learning efficiently and top out at the maximum of their ability. The computer instructor would bring you to a point where you can go no further

We might get to the point where kids learn faster. Gen Z+1 potentially a bunch a little man Tate mother****ers

Not worried, my Valet ChatBot will deal with and run them off.

I think I'm serious.