ChatGPT AI is a huge breakthrough, a Google killer

[WillHurtDontCare]

I'd say use this thread.

IDEA - ChatGPT / AI / MidJourney Creation Repository: Share Your Useful Creations

I am aware there are two or three ChatGPT-related threads on the forum. However, I wanted to make this one separate, to highlight and share random tools and things we create using the AI. Especially for non-coders using it to create neat little things for business, productivity and so on. While...

www.thefastlaneforum.com

I didn't know about this thread - TY, I'll move my post.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[dru-man]

I've tripled my writing output. In fact, I'm now working 4 full-time jobs at once. Won't work for all situations, but I can now take on more and more work. Still, other writers keep leaving their heads in the sand. Thing is, they are also leaving $100k a year on the table.

Would love to hear more on this. Are you doing mostly content work then?

 

[Andy Black]

Interesting prompts he's using:

Ruben Hassid on LinkedIn: How to use ChatGPT | 1,045 comments

Chat GPT. AI revolution. Death of Google. "It will take our jobs!" You kept hearing it at Christmas. To get MORE jobs from it than it takes AWAY, you have… | 1,045 comments on LinkedIn

www.linkedin.com

 

[dru-man]

This thing definitely reveals some interesting biases in its answers (see attached).

It's really bizarre that it would include that last sentence at all -- why would any user assume that describing an individual's ideas equated to supporting them?? Very strong virtue signaling built into the system already...

Asked the bot about other controversial figures that have been around for a while and it acted like it had no knowledge of these individuals, as if that info had been blocked or removed from its original training (even though the bot had no issue describing uncontroversial figures that are roughly just as well-known).

All answers around anything even remotely controversial according to Wokeism also seems to be followed up with a one-liner reminder of how to think politically correct about the issue.

End result seems to be very similar to the way Google slants its search results on certain topics... but with the added danger of the "facts" being fed to you by one single seemingly authoritative source, as opposed to a range of sources giving varying angles (or, in the case of Google results these days, a range of variations that Google deems within the spectrum of acceptability).

Pretty clear how that could become dangerous if it was ever depended on as a single source of information.

Some really interesting stuff and I'm interested to see where it goes, as well as how to leverage it to make money.

Definitely a lot of red flags already though just having played around with it a little.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andy Black]

This thing definitely reveals some interesting biases in its answers.

It's really bizarre that it would include that last sentence at all -- why would any user assume that describing an individual's ideas equated to supporting them?? Very strong virtue signaling built into the system already...

Asked the bot about other controversial figures that have been around for a while and it acted like it had no knowledge of the person, as if that info had been blocked or removed (even though the bot had no issue describing uncontroversial figures that are roughly just as well-known).

All answers around anything even remotely controversial according to Wokeism also seems to be followed up with a one-liner reminder of how to think politically correct about the issue as well.

Yeah. I feel it writes how a lawyer or politician would. That's why I think having a personal opinion and pinning your flag to a mast will become more important if (as?) the internet gets flooded with this kind of speak.

For a few months I was more interested in creating faceless YouTube channels and videos than personal branded talking head ones. Now looking at all the AI that helps people churn out this stuff I'm leaning back towards personal branding and having strong opinions.

 

[dru-man]

Yeah. I feel it writes how a lawyer or politician would. That's why I think having a personal opinion and pinning your flag to a mast will become more important if (as?) the internet gets flooded with this kind of speak.

For a few months I was more interested in creating faceless YouTube channels and videos than personal branded talking head ones. Now looking at all the AI that helps people churn out this stuff I'm leaning back towards personal branding and having strong opinions.

Indeed.

Also... If you consider it largely feeds off whatever's on the internet, even a version capable of scouring current data would potentially suffer from another issue that should be interesting for writers.

Basically, it would still need input sources to work...

So if everyone stopped writing (because of AI supposedly replacing their jobs), its usefulness would quickly get diluted. It would theoretically have less and less original content to draw from, especially on more current topics. In a world where everyone used it to create content (just being hypothetical), it would essentially be recycling its own inputs.

Therefore at least some human content creators would need to be out there creating those inputs to keep it working, right? If this didn't happen by default, then the value of high quality human writing would shoot up in response, if even just for a declining number of well paid specialists.

(I'm certainly no expert on this stuff so very possible I'm missing a lot here. I also realize it will inevitably improve a lot over time. Just thinking outloud.)

My impression as far as writers go, though, would be to either:

1) Become a very specialized kind of writer creating extremely original content (or very opinionated, as you say)... The more nuanced your ideas, the more valuable your skillset as a writer should be -- assuming we don't just all happily fall forward into a world of vanilla filler content reminding us that "it's important to remember to be nice to other people."

(If the last 5 years are any indication, I think humans will naturally push back on that.)

Or 2) Become very skilled at using this technology in high leverage ways so people will pay you big money to use it (this is where I think there are big opportunities right now). Regardless of how it's incorporated into the internet, there are plenty of valuable use cases that anyone who knows how to use it well can start capitalizing on right now.

Another thing I keep thinking about is that even people who use it to create content or copy will need to understand what makes good content or copy. How else would you know enough to edit and improve upon what it produced, or whether it was good enough to use as is?

If everyone depended on the AI, the world's writing skills would suffer as a whole. Hence making top notch writing skills in a human very scarce and therefore (once again) in high demand?

A lot to think about.

I do want to spend some time actually learning how to use this stuff. Hard to say what the future will look like but certainly a lot of economic potential here. And if we can produce amazing things without it, what can we produce by leveraging it?

Crazy to think we're just seeing the beginning.

 

[dru-man]

Also, just want to reiterate what TinyTim said in case all this talk of content creation misleads anyone -- ChatGPT itself should not be used as writing software.

Among other limitations, it often gives back canned answers word for word, and I think others have pointed out that sometimes answers are lifted from sources verbatim. Maybe you can just ask it to use original wording. I haven't played with it much yet.

But just a reminder... you'll probably want to use an AI program actually designed for writing content or copy if you're planning to do something like that.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Wombat]

An interesting use-case for ChatGPT: learning another language:

• Generate natural dialogue at an appropriate level on topics that interest you
• Produce conversations & (with speechify) listen to them
• Explain language concepts (might make things up)
• Create a personalised study plan
• Create language quizzes
• Tell you where you are making a language mistake
• Simplify complex text so you can check you understood it

 

[TinyTim]

Would love to hear more on this. Are you doing mostly content work then?

Yeah, content writing. Most of my work is writing press releases for the same agency. The AI isn't doing all of the work, but it helps to fill in the gaps and give inspiration.

I've had to write about the same company a few hundred times, which as any writer will know, taxes the hell out of my creativity. So I just give the AI a prompt, and then let the AI finish it off. For example, "The use of smart contracts allows X company to...".

I don't find ChatGPT that useful, but I have used it to rewrite some content with - "describe this in a new way".

 

[Andy Black]

Some amazing examples described in this video:

View: https://youtu.be/V-hB-4fnqtM

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andy Bell]

Some amazing examples described in this video:

View: https://youtu.be/V-hB-4fnqtM

Peterson somehow has a way of scaring the hell out of me on every topic. Well then. Is there anything that isn't at risk of being automated by chatgpt? Face to face psychologists maybe and youtube. I mean that's even until dalle moves to video and deepfakes and we have deepfake twerking videos and onlyfans. *hides under bed*

 

[MJ DeMarco]

Some people suck.

https://twitter.com/i/web/status/1611864110244990976

View: https://twitter.com/Austen/status/1611864110244990976?s=20&t=gNDovIcN4kziCsNTXrERJw

 

[MitchC]

Just paid for unbound.ai after seeing their Instagram ads.

Absolute trash, site barely functions, I’d go so far as to call it a scam with how bad it is compared to their marketing.

My opinion after this experience and seeing a dropship video today promoting ai for dropshipping is that while ai can do stuff, it does it badly, and execution is everything so it may as well be useless.

I would say it may replace low level jobs and that I’m wrong for trying to use it for high end creative tasks, but someone in here posted how it couldn’t even do basic maths properly.

You are not going to replace a human with something that does a worse job than them.

If anyone knows of an ai that can actually do what unbound promised let me know, I wanted it to create a model of my product from photos of it and then create lifestyle photos using that model.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andy Black]

Peterson somehow has a way of scaring the hell out of me on every topic. Well then. Is there anything that isn't at risk of being automated by chatgpt? Face to face psychologists maybe and youtube. I mean that's even until dalle moves to video and deepfakes and we have deepfake twerking videos and onlyfans. *hides under bed*

Look at videos on the dreamcloud channel on YouTube. You can get AI to help write a script, then a bot to do the voiceover, and AI to pull scenes from stock images, and AI to add music (or create music from scratch).

There's even AI that creates talking head avatars. There's a channel on TikTok called "AI presents AI" where a talking head character explains different AI tools.

Https://Synthesia.ai might be what it uses.

Btw... everything seems ro be called AI when I think some are just apps.

Here's a other TikTok guy/channel talking about AI:

TikTok - Make Your Day

 

[Andy Black]

Wow. I need to play with ChatGPT more.

Summarising a long video isn't a use case I thought of.

I have a lot of long chats on my YouTube channel that I'd love to summarise.

And there's many long videos on other channels that would be great to summarise.

TikTok - Make Your Day

 

[Ing]

Since 3 days, I can’t log in to chat.openai any more. I am allways redirected to „verify you r a human“.
So at least he recognized, that I m a robot!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Gepi]

Look at videos on the dreamcloud channel on YouTube. You can get AI to help write a script, then a bot to do the voiceover, and AI to pull scenes from stock images, and AI to add music (or create music from scratch).
(...)

That is why good human creators with high quality content will get more sought after. The more things get flooded by AI content, the more a genuine connection with a human being will be craved. At least that's one possible future outlook.

 

[Skroob]

Yeah. I feel it writes how a lawyer or politician would. That's why I think having a personal opinion and pinning your flag to a mast will become more important if (as?) the internet gets flooded with this kind of speak.

For a few months I was more interested in creating faceless YouTube channels and videos than personal branded talking head ones. Now looking at all the AI that helps people churn out this stuff I'm leaning back towards personal branding and having strong opinions.

Put me down as an advocate for personal branding and strong opinions. People connect better with you when you show your face and your feelings. If you don't, there's always a question in your audience's mind. "This could be anybody. Or a bunch of anybodies. How do I know if they're even real?" I've absolutely felt this way about YouTube personalities in the past, and found that my viewership of them waned rapidly.

 

[Andy Black]

Clever:

LINK:

• Nicole Crowell on LinkedIn: #data #chatgpt #research #strategy #frameworks #competitoranalysis | 42 comments

SCREENSHOT:

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andreas Thiel]

Clever:

LINK:

• Nicole Crowell on LinkedIn: #data #chatgpt #research #strategy #frameworks #competitoranalysis | 42 comments

SCREENSHOT:
View attachment 46691

Interesting. Have you played around with that? Given that GPT-3 is not a database I'd intuitively expect that to deliver mixed or subpar results.

 

[Fox]

I find it not that accurate on quite a few things...



Most of this is wrong - and there is a lot of info out there to get the right answer.

So yes it is a powerful tool, but letting this do all the thinking for you is not a smart approach.

 

[Andy Black]

Interesting. Have you played around with that? Given that GPT-3 is not a database I'd intuitively expect that to deliver mixed or subpar results.

I suppose it's kind of a database in that it's scanned 10% of the internet to get trained. The data is just stored differently. I've not tested it though.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[MJ DeMarco]

Well one thing that will stop this from being a "google killer" is every time I open it up to use it, it is unavailable.

 

[ZackerySprague]

Well one thing that will stop this from being a "google killer" is every time I open it up to use it, it is unavailable.

ChatGPT Service Manager: *Where are them Da*mn new servers, we can take this this much heat any longer. We are reaching cap fast!*

 

[Arrived2015]

[...] I feel it writes how a lawyer or politician would. [...]

Hi Andy

LOL

Indeed!

I'm waiting on the outcome with the first UK AI “robot lawyer” case that is due in Feb 2023.
The AI will be listening to the trial, in real-time, before telling the defendant what to say through headphones. The headphones will be considered to be a “hearing aid” .

~ More on this, can be found here:

"World's First Robot Lawyer" To Make History Defending A Client In Court

Don't say a word and listen to the robot.

www.iflscience.com

Also, on 19 February 2019, a ‘robot mediator’ was used to settle a dispute.
It settled the on going three-month dispute in less than an hour!

~ More on this, can be found here

Robot mediator settles first ever court case

A ‘robot mediator’ has been used to settle a dispute in the court system, for what is believed to be the first time. The online tool uses artificial intelligence algorithms in place of a human mediator.

www.legalfutures.co.uk

Then, your politician feeling is not far off because in Oct2022, a humanoid robot became the first machine to speak at a House of Lords committee hearing on AI art.The robot took a long time to answer questions though and at one point even shut down randomly and had to be reset. Ooops!

~ More on this one, found here:

Robot breaks down in middle of House of Lords hearing

Plus: Listen to Fake Joe Rogan interviewing Bogus Steve Jobs in bizarre podcast episode

www.theregister.com

So I'd say trust your feelings because they're spot on

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[heavy_industry]

Peterson somehow has a way of scaring the hell out of me on every topic.

Same for me lol.

ChatGPT is a technology that has been predicted for many decades, and more of its kind will follow. This is how human civilization worked for thousands of years: we build more complex tools that automate manual labor. This usually results in everybody getting richer, so I don't understand what the problem is.

Being scared of an AI system is like being scared of a printing press, factory machine, or vacuum cleaner.

But it's like seeing an airplane or helicopter for the first time: a giant metal beast in the sky that makes horrible noises. Of course it's scary because you don't understand what it is.

 

[dru-man]

One thing I've noticed at least in its current state is that you can get good results with it but have to be somewhat skilled and creative to do so (figuring out the right prompts, digging deeper with further questioning, etc). Anything that requires new skills to be used will probably be abandoned or at least ignored rather quickly by the majority of the population.

Most people just don't have the initiative to learn new tools until they're intuitive. At best, they certainly will fall short of leveraging it's full potential.

I'm often surprised how few people even know how to properly use something as basic as Google maps. Can't tell you how many times I've had a friend ask me where I'm at so they can come meet up with me for a meal or drink or whatever, and I'll forward them a map location, thinking that should solve it... and am left completely baffled that they just keep asking me how to get there, as if they have no idea how to actually use the phone in their hands...

It's just really odd to me. It's like even with today's tech they need the location or directions explained explicitly. And I'm talking about people both younger and older than me, so I don't think it's just a generational thing.

Even finding good answers in Google takes a certain skillset most people have always lacked, especially with all the search engine spam and censorship thrown into the mix. And few have any clue how to tell the difference between fake reviews and real reviews on websites or in Amazon, or how to pick through reviews and other basic research to make an informed online purchase.

Interesting to think about. In a society where we've constantly got our smartphones in our pockets and are always connected, you'd think we'd all be advanced at these basic use skills already. But that's not my experience at all (when dealing with people who don't work predominantly online).

That isn't too say that society won't benefit it from this kind of tech, of course. Just that there will be a major edge to early adopters, entrepreneurs, and "middlemen" that's largely unrealized by the majority...

And for many that will mean profits.

 

[Skroob]

I'm often surprised how few people even know how to properly use something as basic as Google maps. Can't tell you how many times I've had a friend ask me where I'm at so they can come meet up with me for a meal or drink or whatever, and I'll forward them a map location, thinking that should solve it... and am left completely baffled that they just keep asking me how to get there, as if they have no idea how to actually use the phone in their hands...

The reason I'll always have a job if I want it

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andy Bell]

Engadget: The Morning After: Microsoft's VALL-E AI can replicate a voice from a three-second sample.

The Morning After: Microsoft's VALL-E AI can replicate a voice from a three-second sample

The top news stories today: Microsoft's VALL-E AI can replicate a voice from short voice samples, NASA funds ideas and prototypes for future space exploration, The first-ever UK space flight fails to reach orbit.

www.engadget.com

Imagine all the crazy cold calling and spam this would generated. The elderly would have to be banned from using banking without approval

 

[GPM]

Lol, I am trying something here and ChatGPT keeps giving me these woke a$$ disclaimers after its paragraph:

"I would suggest using more carefull wording in your future marketing, as promoting idea that being fit and attractive would make people more successful or happier than others could be seen as harmful to self-esteem, and also some people have disabilities that does not allow them to be "fit" or "attractive" by societal standards."

"Note again that being fit and attractive does not guarantee success or happiness and that promoting such idea could be potentially harmful to some people self-esteem."

What a load of shit. Being and getting fit is 100% a factor in increasing your self esteem and feeling better about yourself. I can feel like an absolute bag of trash, and getting a little bit of exercise in 9/10 times helps me to feel better.

I don't care if you have no legs and are in a wheelchair, you are telling me that that person won't feel better eating a healthy meal and going for a roll and doing whatever they can for exercise to get their heart rate up? Life sometimes sucks, but we are literally living inside flesh bags that need to be physically nourished and exercised in order to run optimally. Should we start advocating that we stop changing out cars oil and tires because it is hurtful to cars that are running on old tires and don't get oil changes?