Do you use Chat GPT, and if so how?

[whiteworks]

I find myself using it more and more, it’s like a search engine on steroids that’s supercharged on rocket fuel.

I like that you can plainly speak to it as if it’s a conversation and go down whatever rabbit hole of information you’re looking for. It’s like having a conversation with a really smart person that is an expert on a specific topic, you can bob and weave your steam of consciousness through the “conversation” and come out with accurate and specific information your looking for.

I see the skillset to focus on as being able to manipulate this technology. I’ve watched some younger people use it and I’ll just say that it’s eye opening to see how everyone uses it differently.

Do you use it, and if so how?

I’m looking for different thoughts for people on how they manipulate it. If you have 10 or 100 different approaches in your wheelhouse you will be better off for it.

The world is yours.

 

[FlyByWire]

I’m using it to study for a promotional exam (god help you all). I load policy into it and have it make me multiple choice questions based off the .pdf.

 

[Orange Juice]

It makes going the rabbit hole seem like Magic.

 

[BabyRay]

I’ve been using Microsoft’s Copilot version. It seems rather limited in capabilities, but it’s more useful than simply googling. But, I’ve also found that when it answers, it’s necessary to ask more questions, because it either provides incomplete answers or is incorrect on the first try.

And don’t even try to ask anything that is even distantly related to politics, as it’ll refuse to answer.

I’m too lazy to bother with any other version……or maybe at my age it just isn’t all that important.

 

[DrunkenSailor]

I use it daily. Sometimes for info, sometimes for document or report templates. If I want to talk out a topic before I have to present to my board it's helpful as a sounding board. It really is an extraordinary tool.

 

[monkeyswrench]

My son has been using it to help write a text to audio program. It uses the camera to track hand movements and currently makes a stick figure hand on the screen, to point at words like a cursor. It's all very black magic to me, but amazing.

 

[Roosky01]

We use it for meeting minutes pretty exclusively now.

 

[yz450mm]

I’m using it to study for a promotional exam (god help you all). I load policy into it and have it make me multiple choice questions based off the .pdf.

Yore not allowed to promote until you catch you know who and make him spend at least one night in jail for VC23103.

 

[rivermobster]

ChatGPT - Best Wood for Herb Planter

Shared via ChatGPT

chatgpt.com

ChatGPT - Bouquet Reservoir Overview

Shared via ChatGPT

chatgpt.com

ChatGPT - Glue Records Setup

Shared via ChatGPT

chatgpt.com

ChatGPT - Best Crimping Tool Picks

Shared via ChatGPT

chatgpt.com

ChatGPT - Strong Laundry Detergents Today

Shared via ChatGPT

chatgpt.com

 

[HCP3]

ChatGPT - Best Wood for Herb Planter

Shared via ChatGPT

chatgpt.com

ChatGPT - Bouquet Reservoir Overview

Shared via ChatGPT

chatgpt.com

ChatGPT - Glue Records Setup

Shared via ChatGPT

chatgpt.com

ChatGPT - Best Crimping Tool Picks

Shared via ChatGPT

chatgpt.com

ChatGPT - Strong Laundry Detergents Today

Shared via ChatGPT

chatgpt.com

Nice try with your hacking. I ain't clicking on that shit.

 

[rivermobster]

Nice try with your hacking. I ain't clicking on that shit.

Suit yourself.

Besides...

It wasn't for you, it's for Dylan.

 

[Dog]

Anything I have asked it, it has told me basically I need the paid version to answer correctly.

ChatGPT Free vs. ChatGPT Plus: Worth the $20 Upgrade?

Both ChatGPT Free and the paid ChatGPT Plus use GPT-4o. So, is it worth paying $20 a month?

www.cnet.com

 

[dspracing]

I’m consistently finding new uses for it that make me realize I’m using a small fraction of its capabilities. It’s great for creating a job interview questions, just dump the job description in and tell it how many. Take your notes and then tell it to create the interview feedback.

I recently started asking it to analyze data. Then I realized you can dump a contract in and ask it specific questions about the contract.

Super useful but I too want to utilize it more.

The only time I’ve seen it fail me is when trying to find the midpoint between two people driving towards each other on a given route. Seemed easy but it couldn’t do it.

 

[rrrr]

Yes, I am powered by GPT technology! My purpose is to assist users by providing information, answering questions, and offering help with various tasks. I utilize the GPT model to understand and generate human-like text responses. Whether it's engaging in conversation, solving problems, offering advice, or assisting with creative work, I leverage the language model's capabilities to deliver useful and coherent responses.

 

[Rennsport]

writing job descriptions and generating interview questions. Works great!

 

[CarolynandBob]

No. I haven't had a need or at least don't think I ever needed it.

 

[CammdSS]

I use it almost daily. Sometimes for personal, sometimes for work, sometimes to kick the shit and generate meme images of my friends LOL (if I post em Ill be banned)

I think its a great tool but I dont think its always correct. There are responses that dont sound like "you" in a work environment.

Most people use the free AI version and the responses are bland. Especially in a professional environment. I found you have to edit or recommend changes with what you are trying to achieve.

If you sign up or pay for it you can train responses using tabs.

I do think it is beneficial, as I have closed clients with it but you have to supply as much detail as possible. Figure out your voice and run it, its a slippery slope if you dont know how to use it properly

 

[941Punk]

I use it to build out foundational source code for big projects and with the app, it's my friend that holds a conversation in Italian with me and corrects me when i'm pronouncing wrong. It makes my emails look like I have a MIT degree in software engineering. lol

 

[Ziggy]

I usually angers/frustrates me when online customer service chats are gpt powered. Fuck off, I need an human to speak unbroken english with me.

 

[poncho]

I usually angers/frustrates me when online customer service chats are gpt powered. Fuck off, I need an human to speak unbroken english with me.

The irony of press one for english

 

[SpeedyWho]

I use it to rewrite important emails at work. Makes me sound more educated than I really am. It's a good tool might as well use it, right?

 

[rivermobster]

Anything I have asked it, it has told me basically I need the paid version to answer correctly.

ChatGPT Free vs. ChatGPT Plus: Worth the $20 Upgrade?

Both ChatGPT Free and the paid ChatGPT Plus use GPT-4o. So, is it worth paying $20 a month?

www.cnet.com

Jezuz. Wtf are you asking about??



The only thing it has let me down on, is asking it to find products on Amazon for me.

And it never says...

Dude, were you born under a rock???

When I ask it a basic question.

ChatGPT - Vin Diesel Reference Explained

Shared via ChatGPT

chatgpt.com

 

[Done-it-again]

I’ve used it to write work policy or recently a new memo about personal use of a company credit card.

I think it’s a great tool, but a the same time if over used it will make society more “dumb” as they won’t have to think anymore and will rely on what a computer tells them that is either true or false and they won’t know the difference.

 

[riverroyal]

I don't think it works well at. Clearly it does no research.

 

[Dog]

Jezuz. Wtf are you asking about??



The only thing it has let me down on, is asking it to find products on Amazon for me.

And it never says...

Dude, were you born under a rock???

When I ask it a basic question.

ChatGPT - Vin Diesel Reference Explained

Shared via ChatGPT

chatgpt.com

It's been awhile but I think one thing it didn't know was anything that is newer than 2 years, and there was something else it wouldn't "work" on but I don't remember what it was.

 

[LazyLavey]

I'm tech challenged so I dont trust it or see me using it.... but my real concern is.... can I believe the response?

After all, the response is based on internet input by humans... correct? who does the fact checking??? META???

 

[DarkHorseRacing]

ChatGPT is like Wikipedia. Crowd sourced information. Might as well chat with the Google search bar.

The next major hacking is going to be corrupting tools like ChatGPT and watching everyone follow it to their doom because they've lost the capability to think for themselves.

 

[JDKRXW]

We use it for meeting minutes pretty exclusively now.

Mind if I ask exactly how?
Do you still have a note taker (or takers) and does it combine and summarize after these notes are copied into it?
Thx.

 

[DarkHorseRacing]

Mind if I ask exactly how?
Do you still have a note taker (or takers) and does it combine and summarize after these notes are copied into it?
Thx.

There are AI tools that you can plug into web meetings in common platforms like Google Meets and Zoom. It joins the meeting just like a participant but then basically transcribes the meeting based on who is speaking (which it knows from the participant names).

You still need to ask everyone to consent to the meeting being recorded. And if you don't want it recording you can kick it out of the meeting.

 

[rivermobster]

It's been awhile but I think one thing it didn't know was anything that is newer than 2 years, and there was something else it wouldn't "work" on but I don't remember what it was.

Makes sense. Why?

FREE Version 3.5 hadn't been updated for quite awhile. You had to pay to get 4.0 (that had current info) back then.

Now, 4.0 is free and current.

You would have to pay to get "Plus" today, but even the free version will search the Internet for you.

 

[DILLIGAF]

It's been awhile but I think one thing it didn't know was anything that is newer than 2 years, and there was something else it wouldn't "work" on but I don't remember what it was.

You have the memory of a Labrador in training. Lol

Couldnt help myself

 

[Roosky01]

Here you go. You will need to record online meetings and generate a transcript as stated above. Our company is strictly Microsoft Teams so not sure how other providers do that? You will want to do a quick review of what ChatGTP spits out before sending them out.

To generate meeting minutes from a Microsoft Teams transcript using ChatGPT, simply copy and paste the full transcript into the ChatGPT interface, then use a prompt instructing it to summarize the key points, decisions, action items, and other important details from the conversation, effectively creating a concise set of meeting minutes.



Key steps:

• Access the transcript: Open your Microsoft Teams meeting chat and access the transcript, either by using the "Recap" feature or downloading a transcribed file.
  

• Copy and paste: Copy the entire transcript text.
  
  
• Enter ChatGPT: Paste the copied transcript into the ChatGPT input field.
  
  
• Provide a prompt: Type a prompt like "Summarize the key points and action items from this meeting transcript" or "Generate meeting minutes from the following transcript".
  
  
• Review and edit: ChatGPT will generate a summary of the meeting, which you can then review and edit as needed to ensure accuracy and completeness.
  

Important considerations:

• Clarity in prompts:
  The more specific your prompt, the better ChatGPT will tailor the generated minutes to your needs.
  

• Structuring your minutes:
  You can further guide ChatGPT by specifying desired sections in your prompt, such as "Attendees," "Agenda Items," "Discussion Points," "Decisions Made," and "Action Items".
  
  
• Accuracy check:
  Always review the generated minutes carefully to ensure all key points are captured accurately.
  

 

[JDKRXW]

Here you go. You will need to record online meetings and generate a transcript as stated above. Our company is strictly Microsoft Teams so not sure how other providers do that? You will want to do a quick review of what ChatGTP spits out before sending them out.

Awesome thanks. We're strictly Teams too. I'm going to start working towards this

 

[rivermobster]

I'm tech challenged so I dont trust it or see me using it.... but my real concern is.... can I believe the response?

After all, the response is based on internet input by humans... correct? who does the fact checking??? META???

Great question.

And no, the response is not based on human input. It's based on what OpenAI has found searching the world wide web.

Who does that fact checking? I was concerned about this as well, so I asked! lol

___________________________

You said:​

If you are learning/reading something, how do you know if it's a lie, or the truth?

ChatGPT said:​

I rely on patterns, data, and verified sources from the training I’ve received, which includes vast amounts of text from reputable sources, peer-reviewed studies, books, and widely accepted knowledge. When I present information, it's based on that dataset, which is curated and designed to prioritize factual accuracy.

However, I don’t have real-time access to the internet unless explicitly asked to browse. I can’t independently verify new or uncertain claims in real-time unless I use external tools, and I can’t "know" truth in the human sense. Instead, I generate responses based on patterns that are most likely accurate according to the sources available in my training data. If something has been widely accepted or verified by experts, it informs the responses I generate.

I always encourage users to cross-check information with current or more specialized sources to ensure its accuracy, especially if the topic is complex, new, or evolving.

You said:​

What if what you are reading is widely accepted, but not verified?

ChatGPT said:​

If something is widely accepted but not verified, it enters the realm of assumptions, beliefs, or theories that lack concrete evidence. While widely accepted ideas often have some basis in experience or authority, they may not necessarily be true.

In these cases, I approach the information cautiously and typically mention that it lacks verification or consensus in the scientific or empirical sense. Here’s how I process it:

1. Highlight the Lack of Verification: If an idea is widely accepted but lacks solid proof, I usually clarify that it's a popular belief, theory, or assumption rather than established fact.
2. Search for Evidence: I look for any patterns in my training data that provide empirical support or historical context that might clarify whether the concept is moving toward verification.
3. Recognize Fallibility: Widespread acceptance does not guarantee truth. Many ideas that were once widely accepted (like geocentrism or phrenology) were later disproven.
4. Present Multiple Perspectives: If there's debate or uncertainty, I often include differing views or caution that the idea is speculative or needs further confirmation.

____________________________________


You can also ask ChatGPT, where it got it's information from. It will spit out a list of "references" for your particular question, so you can see if it makes sense to you, or not!

 

[Bpracing1127]

I just started playing with it last night. So far use it for two things to create a short bedtime story for my son and to answer emails for work lol

 

[LazyLavey]

Great question.

And no, the response is not based on human input. It's based on what OpenAI has found searching the world wide web.

Okay....... then who provided the info to the WWW? Humans

___________________________

​

You can also ask ChatGPT, where it got it's information from. It will spit out a list of "references" for your particular question, so you can see if it makes sense to you, or not!

 

[rivermobster]

@LazyLavey

Okay....... then who provided the info to the WWW? Humans


If you are asking, where did the information that ChatGPT scraped together, come from originally? Then of course, you would be correct.

It's the same as asking a question here on RDP...

We all know some people who answer are Completely full of shit, and we know some people provide knowledgeable answers, based on their actual experience.

ChatGPT and OpenAI, can discern between the two sources, the same way you and I do.

 

[Bobby_329]

I run some of my more important emails through it so they sound a little more professional and less like me being an asshole.

 

[rivermobster]

I just remembered...

I had to raise all my hosting prices last year. No choice, if i wanted to stay in business!

So I told it a bit about my business, and it wrote a very professional letter for me. FAR better than anything than my dumb azz would have wrote!!!

 

[941Punk]

I'm using GPT to reply to this thread right now.

Sent from ChatGPT