ChatGPT

[Pelagic]

2 minutes ago, Eng12345 said:

What models are you guys using to upload files for analysis? 

 

I wanted to play around with that, but chatGPT wants me to pay after ~3 pdf files. 

 

 

 

NotebookLM is probably your best bet for that purpose. I think there might be a limit of around 40ish files per project, but the workaround (check out the sub-reddit on it) is apparently to merge files into one.

 

 

[Eng12345]

28 minutes ago, Pelagic said:

 

NotebookLM is probably your best bet for that purpose. I think there might be a limit of around 40ish files per project, but the workaround (check out the sub-reddit on it) is apparently to merge files into one.

 

 

interesting - thank you. Im playing around with it now, but it is somewhat clunky. 

 

It is helpful though in my quest to not be such a luddite. 

Edited February 8, 2025 by Eng12345

[winjitsu]

In a slightly off topic pivot, I started using Eleven Reader app on my Ipad and Iphone by Eleven Labs to read audio books and other long articles. It's free and a total game changer -- audiobook level narration with interesting voices like Feynman, Burt Reynolds etc. I'm hoping they secured IP deals with estates but not sure...

I found it a bit creepy at first, but love the improvement over robotic AI voices. The advancement in the tech is amazing.

It seems their business model is to offer out these "voices" commercial, like a voice over in a commercial.

[mattee2264]

Saw something on Fin Twit today that got me thinking.....

 

What happens when OpenAI decides to introduce advertising for free accounts? It has over a hundred million monthly active users. That is a lot of eyeballs. People also spend a lot of time on Chat GPT. 

 

Meta and Google have enjoyed a near duopoly over online advertising and it accounts for a sizeable chunk of their revenues. So what happens to these revenues and associated profits when they get competition for advertising dollars from Chat GPT? Especially if Chat GPT decide to price aggressively to attract advertising clients? 

 

Can't be good for profits. 

[formthirteen]

10 hours ago, mattee2264 said:

Meta and Google have enjoyed a near duopoly over online advertising and it accounts for a sizeable chunk of their revenues. So what happens to these revenues and associated profits when they get competition for advertising dollars from Chat GPT?

 

Google loses some of its eyeballs and stops innovating? That already happened. People get a panopticon and operating system in exchange:

 

People Are Using ‘ChatGPT as This Operating System With Their Whole Lives,’ Says OpenAI CEO Sam Altman

https://www.msn.com/en-us/money/technology/people-are-using-chatgpt-as-this-operating-system-with-their-whole-lives-says-openai-ceo-sam-altman/ar-AA1HyiUl

 

Edited June 28, 2025 by formthirteen

[Sweet]

Tech bros think that the world will revolve around their new product.

 

AI is super useful and important, I use ChatGPT a lot, but my google habits haven’t really changed that much.

 

As for as an operating system for life, what exactly are they talking about, or more specifically, who are they talking?  Like maybe a few % will do this, but most won’t.

[formthirteen]

5 hours ago, Sweet said:

Tech bros think that the world will revolve around their new product.

 

Yes. I'm pretty sure Altman is trying to create an operating system. Why wouldn't he? It requires a new generation of users that don't use desktop computers or mobile phones (children). If we got neural implants tomorrow, the new generation would probably be early adopters and most of the previous generations would just laugh at the technology and its users.

 

5 hours ago, Sweet said:

As for as an operating system for life, what exactly are they talking about, or more specifically, who are they talking?  Like maybe a few % will do this, but most won’t.

 

A multimodal (text, voice) interface, that has access to and controls everything you care about.

 

The future is already here, it's just unevenly distributed:

Project Vend: Can Claude run a small shop? (And why does that matter?)
https://www.anthropic.com/research/project-vend-1

 

 

 

 

Spoiler

 

 

Edited June 28, 2025 by formthirteen

[Sweet]

2 hours ago, formthirteen said:

Yes. I'm pretty sure Altman is trying to create an operating system. Why wouldn't he? It requires a new generation of users that don't use desktop computers or mobile phones (children). If we got neural implants tomorrow, the new generation would probably be early adopters and most of the previous generations would just laugh at the technology and its users.

 

Im not saying that he shouldn’t, i’m just saying that those guys live in a bubble, surrounded by other tech bros.

 

I used ChatGPT daily, some of my friends do too.  But my family, and many others either use it rarely, or not at all, and really have no need or desire to use a ‘life operating system’.

 

I’m saying that these guys don’t understand the normie mindset, which is  a comfortable life, nice food, nice holidays etc.  Its not syncing their life with an operating system.

 

There is a whole section of women who for fun enjoy scrolling Instagram or Pinterest for fashion, food or home tips.  I don’t get it either but the idea that these habits are going to ChatGPT I think is unlikely.

[formthirteen]

1 hour ago, Sweet said:

I’m saying that these guys don’t understand the normie mindset, which is  a comfortable life, nice food, nice holidays etc.  Its not syncing their life with an operating system.

 

Yes, it's a bubble and the bubble will be the future for many normies, particularly the new generation.

 

"syncing their life with an operating system" is automatic when most of the things a normie does is through the operating system (ChatGPT et al). I guess that's the idea Altman might be trying to sell.

 

ChatGPT already provides "assisted-living" services to normies while being connected to their life, whatever that means:



 

 

That query saves people more time than the "old trusty" Google.

[Sweet]

2 hours ago, formthirteen said:

 

Yes, it's a bubble and the bubble will be the future for many normies, particularly the new generation.

 

"syncing their life with an operating system" is automatic when most of the things a normie does is through the operating system (ChatGPT et al). I guess that's the idea Altman might be trying to sell.

 

ChatGPT already provides "assisted-living" services to normies while being connected to their life, whatever that means:



 

 

That query saves people more time than the "old trusty" Google.

 

I doubt many are choosing their restaurants based on suggestions from ChatGPT.

 

I still use trusty old Google if I am checking out the reviews from local food places.  I’ve never once used ChatGPT for that.

 

What I use ChatGPT is for knowledge based questions, the responses are good, but the Google AI responses are good too - and in some cases better.

 

Example.  I’m not a DIY type person, but I have a hole in the wall around a pipe from a recent building work done to the house.  I asked ChatGPT how to fix it, but it didn’t seem right and I check with my father and online forums and they suggested something else.  I happened to google it and the Google AI overview gave a better guide on how to fix the hole than ChatGPT.  
 

So what’s happening for me is that some of these knowledge searches are actually being checked on both Google and ChatGPT now.

 

Edit - my intention was not to do a ChatGPT vs Google argument, rather than there is a large section of society that don’t give a crap about AI at all and for them that ‘life servers’ crap is just out of touch.

 

Edited June 28, 2025 by Sweet

[formthirteen]

39 minutes ago, Sweet said:

Edit - my intention was not to do a ChatGPT vs Google argument, rather than there is a large section of society that don’t give a crap about AI at all and for them that ‘life servers’ crap is just out of touch.

 

Yes, you have a valid point of view. My point is that Altman et al. are aiming at putting their own service (operating system) in front of not just search, as Google has done, but also as many other services and tools as they can.

 

Cloudflare's CEO shares some insights on what is happening behind the scenes:

Bots are automating web search (and hiding the original content in the process):
 

Quote

- 10 years ago: Google crawled 2 pages per visitor
- 6 months ago: Google 6:1, OpenAI 250:1, Anthropic 6,000:1
- Now: Google 18:1, OpenAI 1,500:1, Anthropic 60,000:1

 

This could also be why people on the forum are seeing Cloudflare's bot prevention on the COBF forum.

[Spekulatius]

1 hour ago, Sweet said:

 

I doubt many are choosing their restaurants based on suggestions from ChatGPT.

 

I still use trusty old Google if I am checking out the reviews from local food places.  I’ve never once used ChatGPT for that.

 

What I use ChatGPT is for knowledge based questions, the responses are good, but the Google AI responses are good too - and in some cases better.

 

Example.  I’m not a DIY type person, but I have a hole in the wall around a pipe from a recent building work done to the house.  I asked ChatGPT how to fix it, but it didn’t seem right and I check with my father and online forums and they suggested something else.  I happened to google it and the Google AI overview gave a better guide on how to fix the hole than ChatGPT.  
 

So what’s happening for me is that some of these knowledge searches are actually being checked on both Google and ChatGPT now.

 

Edit - my intention was not to do a ChatGPT vs Google argument, rather than there is a large section of society that don’t give a crap about AI at all and for them that ‘life servers’ crap is just out of touch.

 

You definitely want to use different AI bots for any serious work. I’d have found answers from Gemini and ChatGPT the most useful. Sometimes, the answers are contradictory so you need to pick what makes the most sense or try both.

 

I think a better way to describe wha AI Chatbots do is coaches rather than operating systems. You can ask them for ides and then try to expand on them using other source. If you can bounce of your ideas and see if they are sense or ask them for sources. If they don’t have any or not credible ones, thread carefully.

 

About a year ago, LLM Chatbots  were totally useless for my work domain, but now they seem to ingested enough data to come up with useful ideas on well targeted inquiries. The progress that is made here is extraordinary, imo.

[Sweet]

34 minutes ago, Spekulatius said:

You definitely want to use different AI bots for any serious work. I’d have found answers from Gemini and ChatGPT the most useful. Sometimes, the answers are contradictory so you need to pick what makes the most sense or try both.

 

I think a better way to describe wha AI Chatbots do is coaches rather than operating systems. You can ask them for ides and then try to expand on them using other source. If you can bounce of your ideas and see if they are sense or ask them for sources. If they don’t have any or not credible ones, thread carefully.

 

About a year ago, LLM Chatbots  were totally useless for my work domain, but now they seem to ingested enough data to come up with useful ideas on well targeted inquiries. The progress that is made here is extraordinary, imo.

I had to write a report for the area I work in, science basically.  The output was generic garbage.  Overall I think using LLMs cost me more time because I tried to work with the output but ended up just binning it.  I wonder if - when it will make a big impact to my work.  In theory it should but right now it can’t.