Asking google AI the right question

[chalkoutline]

Ive been studying Google AI for a while to understand the benefits and quirks. My method of research is to conduct my own tests and record the responses it is done in isolation as much as possible to to avoid bias and outside influence.
Rule #1: be very, very specific.
Rule #2: don't stop when you get the answer you want.

Here is an example...
Question 1: Can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft in 1600s?
Answer: Yes, it was possible to dig a 100ft vertical shaft in clayish till with log platforms every 10ft in the 1600s, using manual tools like picks, shovels, and windlasses. Clayish till is firm, allowing for structural stability.

Question 2: can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft WITHOUT WALL SUPPORT in 1600s
Answer: Digging a 100-foot vertical shaft in clayish till in the 1600s is theoretically possible but highly dangerous and improbable without significant, continuous, and comprehensive vertical wall support (timbering or cribbing).

You can see from the answer to question 1, that Google AI ASSUMED the use of wall support.
Once the "no wall support" is added to question 2, the answer is completely different.
-CO

 

[chalkoutline]

Ive been studying Google AI for a while to understand the benefits and quirks. My method of research is to conduct my own tests and record the responses it is done in isolation as much as possible to to avoid bias and outside influence.
Rule #1: be very, very specific.
Rule #2: don't stop when you get the answer you want.

Here is an example...
Question 1: Can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft in 1600s?
Answer: Yes, it was possible to dig a 100ft vertical shaft in clayish till with log platforms every 10ft in the 1600s, using manual tools like picks, shovels, and windlasses. Clayish till is firm, allowing for structural stability.

Question 2: can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft WITHOUT WALL SUPPORT in 1600s
Answer: Digging a 100-foot vertical shaft in clayish till in the 1600s is theoretically possible but highly dangerous and improbable without significant, continuous, and comprehensive vertical wall support (timbering or cribbing).

You can see from the answer to question 1, that Google AI ASSUMED the use of wall support.
Once the "no wall support" is added to question 2, the answer is completely different.
-CO

Rule #3: Always check the sources used by Google AI...Is it a published document, website, forum, etc.

You will find that Google AI will use responses from Treasurenet as its source and often they are just opinions so you MUST read the source.

I have actually found Google AI used a post of mine post on the Beale treasure as a source. Of course that one was correct

 

[AusTexDude]

I have been using Chat GPT more advanced version. Mind is blown. I can go into a thrift store, take a picture of an entire shelf, say "Price Please" and it will identify everything on the shelf, give me a buy price for each one, what it sells for on ebay, tell me to avoid certain things, and why. I'm just like wow.

 

[chalkoutline]

Rule #4: Beware of convergence!
Google AI is always learning and answers change during the learning process. It's not until it has reached "convergence" that you get consistent, accurate answers.

So it is very important to ask the question differently multiple times to verify "convergence" has occurred before using the answer in a post.

 

[AusTexDude]

Look at this amazing description it wrote me for one of my eBay sales.

This isn’t racing merchandise.
It’s insider memorabilia.

What you’re looking at is an official COTA Lone Star Le Mans promotional leather wine case, produced not for sale, not for souvenir booths, and not for fans in the grandstands. This was created for VIPs, sponsors, executives, and insiders connected to the Lone Star Le Mans at Circuit of the Americas.

Two bottles.
Heavy leather.
Purpose-built.

Items like this exist in a very small window:

• After contracts are signed
• Before the public ever sees anything
• When brands are speaking directly to the people who matter

That’s why these don’t show up in gift shops. They’re handed out quietly, used briefly, and then disappear into private collections — or never leave the room at all.

The design tells the story:

• Racing heritage without being loud
• Luxury without logos screaming for attention
• The kind of piece meant to sit in a hospitality suite, not a merch shelf

This is the sort of object that gets passed across a table after a long day at the track, while deals are closed and champagne (or something better) is poured.

If you collect motorsport memorabilia, this sits in a different category entirely:

• Not mass-produced
• Not reissued
• Not replaceable

Once these are gone, they’re gone.

You don’t buy this because you like racing.
You buy this because you understand access, provenance, and the difference between fan gear and artifacts.

This is Le Mans culture, filtered through American motorsport, wrapped in leather, and never intended to be listed online.

And now it is.

 

[chalkoutline]

Rule #5: Be leary of those using Google AI in a post
If someone has a new theory and what to use Google AI as support of that theory, take it with a grain of salt since they may have asked an incomplete question to get the answer they wanted.

That goes for my posts as well...please fact check me!

BTW, these rules are based upon my own research and observations.

 

[chalkoutline]

Here is another example: Cornish Mining Pick

Question 1: Who found a Cornish mining pick in MP on oak island?
Answer: Gary Drayton and Jack Begley discovered an antique pickaxe, believed to be a Cornish mining pick, at Isaac’s Point on Oak Island

Question 2: Who found a Cornish mining pickwhile digging a shaft on oak island?
Answer: Based on accounts of the 1931 excavations at Oak Island, William Chappell identified the pickaxe found at the 127-foot level of his shaft as a "Cornish miner's pick" based on its specific design, which was characteristic of hard-rock mining tools used in Cornwall, England, during the 18th and 19th centuries.

The sources at the bottom revealed...
The County of Cornwall is the westernmost part of UK’s south-west peninsula, and is part of the Cornubian Orefield which covers Cornwall and some of Devon. Mining began in the early Bronze Age, approximately 2150 BC

This information prompts question 3: When was the cornish mining pick invented?
Answer: The earliest evidence of mining picks in Cornwall dates back to the Early Bronze Age, approximately 2400-800BC.

Double check the AI response with additionaldetail...
Question 4: When was the cornish mining pick WITH NARROW DESIGN FOR DIGGING INTO ROCK invented?
Answer: The earliest specialized, narrow-designed Cornish mining picks—specifically fashioned for cutting into rock—date back to the Early Bronze Age, around 2000–2100 BC.

My take aways are...
1. William Chappell identified the Cornish pick by shape/design and not through metrology.
2. The design had been around for 4000 years so the design would have been copied and used worldwide.

My conclusion...although the pick was based on a Cornish design, no one can definitively conclude it was left by Cornish miners while digging in the Money Pit.

 

[chalkoutline]

Here is another example: Cornish Mining Pick

Question 1: Who found a Cornish mining pick in MP on oak island?
Answer: Gary Drayton and Jack Begley discovered an antique pickaxe, believed to be a Cornish mining pick, at Isaac’s Point on Oak Island

Question 2: Who found a Cornish mining pickwhile digging a shaft on oak island?
Answer: Based on accounts of the 1931 excavations at Oak Island, William Chappell identified the pickaxe found at the 127-foot level of his shaft as a "Cornish miner's pick" based on its specific design, which was characteristic of hard-rock mining tools used in Cornwall, England, during the 18th and 19th centuries.

The sources at the bottom revealed...
The County of Cornwall is the westernmost part of UK’s south-west peninsula, and is part of the Cornubian Orefield which covers Cornwall and some of Devon. Mining began in the early Bronze Age, approximately 2150 BC

This information prompts question 3: When was the cornish mining pick invented?
Answer: The earliest evidence of mining picks in Cornwall dates back to the Early Bronze Age, approximately 2400-800BC.

Double check the AI response with additionaldetail...
Question 4: When was the cornish mining pick WITH NARROW DESIGN FOR DIGGING INTO ROCK invented?
Answer: The earliest specialized, narrow-designed Cornish mining picks—specifically fashioned for cutting into rock—date back to the Early Bronze Age, around 2000–2100 BC.

My take aways are...
1. William Chappell identified the Cornish pick by shape/design and not through metrology.
2. The design had been around for 4000 years so the design would have been copied and used worldwide.

My conclusion...although the pick was based on a Cornish design, no one can definitively conclude it was left by Cornish miners while digging in the Money Pit.

I asked Google AI one more question...
Question 5: What is the probability that the pick found by William Chappell at 127ft was left by Cornish miners on oak island?
Answer: Based on historical accounts, the probability that the pick found by William Chappell at 127 feet in 1931 was left by original, ancient Cornish miners is low to unverifiable

 

[robertk]

Rule #3: Always check the sources used by Google AI...Is it a published document, website, forum, etc.

You will find that Google AI will use responses from Treasurenet as its source and often they are just opinions so you MUST read the source.

And that, in my opinion, is the biggest flaw in AI. It’s not intelligence, it’s a search engine with prioritization and inference. It’s only as good as its sources. And we all know how some sources get favored while others get buried. And even the absolute best sources are sometimes wrong.

 

[Armchair detective]

So ... treat it like another opinion instead of an oracle.

 

[Crow]

Raggedy old Crow, who is still awaiting his bourbon, a secret. Not all information and knowledge is published online. Archives and books still have relevance. While the internet is revolutionary, AI has come a long way. It's not infallible.

Crow

 

[pepperj]

And that, in my opinion, is the biggest flaw in AI. It’s not intelligence, it’s a search engine with prioritization and inference. It’s only as good as its sources. And we all know how some sources get favored while others get buried. And even the absolute best sources are sometimes wrong.

I agree that it's selective in its searches. Who ever has the risen to the top tier of searches it will gleam those first.
It can be manipulated into giving an answer that one wants actually.
Not to be completely trusted in certain areas of information gathering either.
It can be very flawed in calculations of what something can weigh even.
It's at a 60% level I would say. It passes but needs alot of improvements.

In 10 yrs it will be producing this series with them extracting the treasures. Nobody will ever doubt that it didn't happen.
That's the scary part, folks believe what they see on the screen. I saw it on YOUTUBE so it's the truth.

 

[tamrock]

I have been using Chat GPT more advanced version. Mind is blown. I can go into a thrift store, take a picture of an entire shelf, say "Price Please" and it will identify everything on the shelf, give me a buy price for each one, what it sells for on ebay, tell me to avoid certain things, and why. I'm just like wow.

No kidding. I'll have to try that. Quite an advancement from the days of scanning bar codes on thrift store books to see what they might sell for. I've been having long drawn out conventions with my echo dot. Its great for coming up with dinner ideas after asking it for advice on what to make with what you already have on hand.

 

[AusTexDude]

No kidding. I'll have to try that. Quite an advancement from the days of scanning bar codes on thrift store books to see what they might sell for. I've been having long drawn out conventions with my echo dot. Its great for coming up with dinner ideas after asking it for advice on what to make with what you already have on hand.

It will not do that automatically. I had to set up Hal (thats what I named him) to do that. It's easy, I just told him to remember when I type in "Price Please" and send a pic of items to do all that. I also have it set up so when I send it a picture and say Ebay Listing, it will build me an entire ebay listing, even telling me how many OZ to add to the shipping weight to account for box and materials. Because of Hal I have been able to list almost 200 things in less than 30 days.

 

[tamrock]

It will not do that automatically. I had to set up Hal (thats what I named him) to do that. It's easy, I just told him to remember when I type in "Price Please" and send a pic of items to do all that. I also have it set up so when I send it a picture and say Ebay Listing, it will build me an entire ebay listing, even telling me how many OZ to add to the shipping weight to account for box and materials. Because of Hal I have been able to list almost 200 things in less than 30 days.

wow!

 

[Oak Island Sage]

Ive been studying Google AI for a while to understand the benefits and quirks. My method of research is to conduct my own tests and record the responses it is done in isolation as much as possible to to avoid bias and outside influence.
Rule #1: be very, very specific.
Rule #2: don't stop when you get the answer you want.

Here is an example...
Question 1: Can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft in 1600s?
Answer: Yes, it was possible to dig a 100ft vertical shaft in clayish till with log platforms every 10ft in the 1600s, using manual tools like picks, shovels, and windlasses. Clayish till is firm, allowing for structural stability.

Question 2: can you dig a 100ft vertical shaft in clayish till placing log platforms every 10ft WITHOUT WALL SUPPORT in 1600s
Answer: Digging a 100-foot vertical shaft in clayish till in the 1600s is theoretically possible but highly dangerous and improbable without significant, continuous, and comprehensive vertical wall support (timbering or cribbing).

You can see from the answer to question 1, that Google AI ASSUMED the use of wall support.
Once the "no wall support" is added to question 2, the answer is completely different.
-CO

Hi chalkoutline, the only difference is that in answer 2/ it says that again it is "theoretically possible but highly dangerous". Hazardous or not so it remains a distinct likelihood that experienced Cornish Miners did not find the need to use cribbing, and they were renown for their mining expertise.

They probably used different shape picks for excavating the walls of vertical shafts, and smaller short handled ones for digging flood tunnels that were said to be approximately 4ft high by 2ft wide, there were many smaller people in those days. Your objection for question 2 seems to hinge on whether or not the miners were brave enough not to use cribbing, how the heck are you going establish that? you can only speculate like the rest of us because you were not there either.
Personally I believe the old records for one simple reason, logically speaking what could they possibly gain by lying about the discovery or lack of cribbing. Apart from you who else really cares one way or the other, most people just want to see the treasure- period.
Oak Island Sage

 

[robertk]

They probably used different shape picks for excavating the walls of vertical shafts, and smaller short handled ones for digging flood tunnels that were said to be approximately 4ft high by 2ft wide, there were many smaller people in those days.

They'd have to be really short. That or dig the tunnels lying down. The average height of European men in the 1700s was 5' 5", so they couldn't stand up in a 4' tunnel, much less swing a pick.

 

[chalkoutline]

Hi chalkoutline, the only difference is that in answer 2/ it says that again it is "theoretically possible but highly dangerous". Hazardous or not so it remains a distinct likelihood that experienced Cornish Miners did not find the need to use cribbing, and they were renown for their mining expertise.

They probably used different shape picks for excavating the walls of vertical shafts, and smaller short handled ones for digging flood tunnels that were said to be approximately 4ft high by 2ft wide, there were many smaller people in those days. Your objection for question 2 seems to hinge on whether or not the miners were brave enough not to use cribbing, how the heck are you going establish that? you can only speculate like the rest of us because you were not there either.
Personally I believe the old records for one simple reason, logically speaking what could they possibly gain by lying about the discovery or lack of cribbing. Apart from you who else really cares one way or the other, most people just want to see the treasure- period.
Oak Island Sage

You always selectively edit the facts to fit your narrative...

theoretically possible but highly dangerous and IMPROBABLE WITHOUT SIGNIFICANT, CONTINUOUS AND COMPREHENSIVE VERTICAL WALL SUPPORT (TIMBERING OR CRIBBING).

Who can't handle the truth now?

 

[russau]

Another point thinking along these lines is : The Government Agency's (BLM , Forest Service, Wacoenviromentalists ETC. ) have been using this method of taking comments off line to use in the courts for a LONG TIME ! And "OF COURSE IT'S TRUE ! It was on the internet " I've read this on the net , and no I was not there to hear it ! Be careful of what you put in type over the net , It will come back ti bite you ! Jerry Hobbs of PLP reported that bit of info !

 

[chalkoutline]

Your objection for question 2 seems to hinge on whether or not the miners were brave enough not to use cribbing, how the heck are you going establish that? y

OIS.
As you mention over and over that Cornish miners were very skilled at what they do.
Based on their expertise handed down generations it seems they would have known digging a vertical shaft without wall support would be highly dangerous AND IMPROBABLE.
If they didn't know this, then they were not as experienced and skilled as you suggest.

The first thing a skilled miner would do is take a sample of the dirt. Using his knowledge of previous shafts, i suggest they would say...
"hmmm, clayish till, not a good sign. This isn't the stiff clay like we have in Cornwall so we probably need to use wall support to be safe."

Or as you suggest, they would say...
"hmmm, clayish till. Well, we know how to use cribbing when digging vertical shafts but thats only when the crew are a bunch of cowards. However, this crew is really, really brave, so damn the cribbing! We're NOT going to use any support to prevent cave ins until we get to 100ft mark where we start digging horizontal tunnels.. then we'll use cribbing!"