I asked ChatGTP to suggest a withdrawal strategy.

2HOTinPHX

Full time employment: Posting here.
Joined
Mar 1, 2015
Messages
981
Location
Phoenix AZ
Sorry for starting another ChatGTP chat thread it was very long so deleted and added a summary per request in another reply below. It has been expressed that not many here are interested in long detailed conversations with ChatGTP. I find the detail and where the conversations go amazing. I thought others would want to see the thought progression of AI to but alas I was wrong. My bad, won't happen again.
It's what computers were meant to be from the start IMO. If you haven't tried it I suggest you do. Google search ain't got nothing on ChatGTP.... 😉

Screenshot_2025-01-02-14-42-35-46_40deb401b9ffe8e1df2f1cc5ba480b12.jpg

Picture generated by GTP!!!!!
 
Last edited:
Pretty impressive!

My understanding of guardrails is a bit different but there are many variations on that concept.
 
Copy the wall of words and ask chat to summarize in one page?

I heard something interesting on a podcast recently. They were discussing chat hallucinations, and remarked that hallucinating is the way these engines work. It's just that most times the output seems reasonable to us. Food for thought.
 
Honestly, @2HOTinPHX, your continued posting of these long dialogs with ChatGPT is getting kind of old. The forum is for people to interact with each other.

We have rules against copy/pasting long sections from copyrighted material, and while we don't have any specific rule about AI material, I think we would all appreciate it if you would just briefly summarize these things instead of posting them. We'd like to hear your views, not the AI's.
 
Honestly, @2HOTinPHX, your continued posting of these long dialogs with ChatGPT is getting kind of old. The forum is for people to interact with each other.

We have rules against copy/pasting long sections from copyrighted material, and while we don't have any specific rule about AI material, I think we would all appreciate it if you would just briefly summarize these things instead of posting them. We'd like to hear your views, not the AI's.
Wow... sorry... noted it won't happen again. I thought peeps here would be interested in the actual conversations ChatGTP can develope and the journey it takes us on. I am amazed at it's detailed info and I thought others would be to. I guess I was wrong. It won't happen again. You can delete all the ChatGTP threads I started so others don't waste there time.
I now return you all to the longest thread in the history of threads. Please continue the never ending discussion on when to take social security.... 😉
 
Not a problem. The stuff is interesting, but it's not what members here expect. Your own opinions are more important, since an AI will give a different answer to the next person who asks the same question.
 
That's cool I will keep things here brief and limited to my opinion when I think I can add to an ongoing conversation.
I will ask CHATGTP to find us a new forum that would be more compatible for long winded rewrites of The Sound of Music. We already have concepts of a plan for Gone with the Wind set on Mars..... 😜
Live long and prosper...


Screenshot_2025-01-02-14-20-35-05_680d03679600f7af0b4c700c6b270fe7.jpg
 
Last edited:
Wow... sorry... noted it won't happen again. I thought peeps here would be interested in the actual conversations ChatGTP can develope and the journey it takes us on. I am amazed at it's detailed info and I thought others would be to. I guess I was wrong. It won't happen again. You can delete all the ChatGTP threads I started so others don't waste there time.
I now return you all to the longest thread in the history of threads. Please continue the never ending discussion on when to take social security.... 😉
It's interesting, but you can share a link to all of that maybe, and summarize.
 
It's interesting, but you can share a link to all of that maybe, and summarize.
Thanks for the reply. I will consider that going forward. I know they were long but I was just trying show the detail and complexity of the way AI can take a simple idea or question and really turn it in to something amazing. It really is like the human brain having access to all the knowledge out there and in a matter of seconds coming up with ideas or potential solutions and out of the box thinking. Like the human brain it can make mistakes but also human like is it's ability to show a sense of humor along the way.
Screenshot_2025-01-02-14-42-35-46_40deb401b9ffe8e1df2f1cc5ba480b12.jpg
 
Ok so here is the summary some requestwd..it loses a lot of the details. I guess it's like going to a Financial Advisor and him handing you a one page plan. Details we don't need no stinking details.... 😉 Anyways back to wordel....

Retirement Investment Strategy Overview
Portfolio Allocation

50/50 Stocks-to-Bonds Split: Balances growth and stability.
Stocks (50%): Large-cap dividends, U.S. index funds, and 20–30% international exposure.
Bonds (50%): TIPS, high-quality corporates, and municipal bonds.
Dynamic Withdrawal Strategy

Start with a 4% annual withdrawal rate (~$44,000/year).
Adjust spending using guardrails: Increase withdrawals if the portfolio grows by 20% or more, and reduce them if it drops by 20% or more.
Long-Term Care and Charitable Giving

Reserve assets for potential care needs.
Focus on charitable giving after passing, using tools like living trusts and beneficiary designations.
Advanced Guardrail Example

Portfolio Thresholds: $1.32M (upper) and $880K (lower).
Spending adjusts within these limits to preserve funds and reduce risk.
Recommended Tools

NewRetirement: In-depth planning platform.
Personal Capital/Empower: Real-time financial tracking.
RightCapital or i-ORP: Advanced tax and withdrawal optimization.
 
Ok so here is the summary some requestwd..it loses a lot of the details. I guess it's like going to a Financial Advisor and him handing you a one page plan. Details we don't need no stinking details.... 😉 Anyways back to wordel....

Retirement Investment Strategy Overview
Portfolio Allocation

50/50 Stocks-to-Bonds Split: Balances growth and stability.
Stocks (50%): Large-cap dividends, U.S. index funds, and 20–30% international exposure.
Bonds (50%): TIPS, high-quality corporates, and municipal bonds.
Dynamic Withdrawal Strategy

Start with a 4% annual withdrawal rate (~$44,000/year).
Adjust spending using guardrails: Increase withdrawals if the portfolio grows by 20% or more, and reduce them if it drops by 20% or more.
Long-Term Care and Charitable Giving

Reserve assets for potential care needs.
Focus on charitable giving after passing, using tools like living trusts and beneficiary designations.
Advanced Guardrail Example

Portfolio Thresholds: $1.32M (upper) and $880K (lower).
Spending adjusts within these limits to preserve funds and reduce risk.
Recommended Tools

NewRetirement: In-depth planning platform.
Personal Capital/Empower: Real-time financial tracking.
RightCapital or i-ORP: Advanced tax and withdrawal optimization.
Is this from GPT plus? Haven't tried this query but I'd be interested in giving it a go.
 
Is this from GPT plus? Haven't tried this query but I'd be interested in giving it a go.
No its just the basic non paid version. The detail it went into was amazing. Good luck with your chat.
 
Last edited:
Do we find out exactly what prompt you started with?
 
Is this from GPT plus? Haven't tried this query but I'd be interested in giving it a go.
Try a prompt like this: Let's talk investment strategy for our retirement.
The corrected link added in below post.
 
Last edited:
Is this from GPT plus? Haven't tried this query but I'd be interested in giving it a go.
Ok I think this link will work for those interested in a VERY deep dive into the mind of ChatGPT. The retirement planning it suggested to me was based on some similar chats with it on retirement subjects previously. It has some sort of memory use going on. Maybe this conversation will give you ideas to discuss with it for your own situation.

 
RightCapital or i-ORP: Advanced tax and withdrawal optimization.

I think this is one reason people question the value of AI at this point. I-ORP has been defunct for quite a while.

I find your AI posts interesting. Maybe someday I'll have to play with AI.
 
Yes it is not perfect but if you question it or point out an error it will research and correct usually with an apology. I won't post the response to your error here but it did acknowledge the error and included a link to a thread here discussing the IORB issue.

Give it shot next time you want to do a Google search. I find most answers better than wading thru all the Google links. Just Google CHATGTP and select it. And hit start now....you will be amazed the way you can ask follow up questions... 😉
 
Last edited:
I’m sensitive to this because I have a friend who’s stopped thinking for himself, and he replies with lengthy ChatGPT responses, it’s was interesting at first but now it’s just tiresome. By (subconsciously) structuring his questions to give him the answer he wants, he leaves counterpoint to the rest of us. He has the answer he wants, and doesn’t think any further.

I’d rather hear what another person thinks instead of hearing what ChatGPT condenses from the internet. ChatGPT isn’t immune to misinformation…
 
I’m sensitive to this because I have a friend who’s stopped thinking for himself, and he replies with lengthy ChatGPT responses, it’s was interesting at first but now it’s just tiresome. By (subconsciously) structuring his questions to give him the answer he wants, he leaves counterpoint to the rest of us. He has the answer he wants, and doesn’t think any further.

I’d rather hear what another person thinks instead of hearing what ChatGPT condenses from the internet. ChatGPT isn’t immune to misinformation…
Good points indeed. Sorry you have to deal with a friend like that. Perhaps CHATGPT can help draft a letter of concern to him....joke..
I admit I am a little bit like a kid in the candy store who just discovered this AI stuff. I was just trying to share something I found amazing and different. It was overboard for some with lengthy AI post so I won't be doing that anymore.

Yes it isn't immune to misinformation just like humans. I say use it as a source not the source. I to have found errors in some of its responses. It is learning though....and it's here to stay for better or worse ... LOL
Screenshot_2025-01-02-19-14-05-35_680d03679600f7af0b4c700c6b270fe7.jpg
 
Last edited:
Ok I think this link will work for those interested in a VERY deep dive into the mind of ChatGPT. The retirement planning it suggested to me was based on some similar chats with it on retirement subjects previously. It has some sort of memory use going on. Maybe this conversation will give you ideas to discuss with it for your own situation.

Ok, I get the gist of your interaction now.

It appears to me as a dry description of various artifacts in its store. It gives you a structure you can follow up on, discuss with others, and personalize.

Because of the length and depth, it is like giving a thirsty traveller a fire hose when the ask for a drink.

So it's better than just a web search, but you still need a review of all of that before taking action.

What happens if you use something like PerplexityAI, which can use any one of several engines, and also uses linked web sources in the output?

And as we know after inspection, this is a good start, but adding peers to the follow-up discussion probably can lead to an actionable plan, or some guidance as required.
 
Ok, I get the gist of your interaction now.

It appears to me as a dry description of various artifacts in its store. It gives you a structure you can follow up on, discuss with others, and personalize.

Because of the length and depth, it is like giving a thirsty traveller a fire hose when the ask for a drink.

So it's better than just a web search, but you still need a review of all of that before taking action.

What happens if you use something like PerplexityAI, which can use any one of several engines, and also uses linked web sources in the output?

And as we know after inspection, this is a good start, but adding peers to the follow-up discussion probably can lead to an actionable plan, or some guidance as required.
That's what I was attempting to do have some of the knowledgeable people give some feedback but apparently many here we're not interested in a deep dive. So I deleted the rather lengthy original post. The summary created left a lot of important details out unfortunately.I guess that's to be expected with a subject like retirement with so many variables and options and opinions.

Definitely covered the basics for a good plan I thought. A good starting point but not the end all. But yes you are correct continue to do your own research and find what works for you and let's you sleep at night. After all it's a guessing game as nobody can predict what's going to happen next not even AI... 😉
 
I think the problem here is that many posters are very experienced and we’ve had conversations about withdrawal strategies many, many times over the years, and even recently, counseling newcomers. So no, no one wants to wade through the extended details from ChatGPT because so much of it reads like same old, same old regurgitated stuff.
 
Last edited:
Back
Top Bottom