You're so full of BS, but don't even know it. Please study LLMs; you are assuming way too much about ChatGPT/LLMs.
You have to be smart enough/knowledgeable enough to fact check ChatGPT every step of the way. I just asked it for very basic info on three related companies and it spat out bullet points and had the locations wrong on two of them...not even close. I questioned and corrected it and it went out verified and agreed with the correction. When I asked why it got the locations wrong it simply replied:
"You're right - that is a basic fact, and I got it wrong.
What happened: I answered from my internal memory/assumptions instead of verifying the locations first."
@Taghkanic
I use AI all the time for work to analyze web pages/spec sheets, crunch data, etc. Even these very concrete tasks have to be massaged and corrected.
Here is a final analysis of this simple stupid example from above regarding physical locations and production methods of some coffee mugs. I asked ChatGPT to summarize it's failures:
Please summarize your methodology and faults in determining locations and manufacturing processes for these 3 vendors.
That’s a fair request. Here is a clear, honest post-mortem of how I reasoned, where it failed, and what should have been done instead—without hand-waving.

ne: Methodology I Used (and Why It Was Flawed)
A) Pattern inference instead of verification
I relied on industry pattern matching rather than source validation.
In custom drinkware, there is a common pattern:
Tourist / branded mugs at scale → often slip-cast
Wheel-thrown → often smaller studios, less uniform output
I incorrectly applied that general heuristic to specific companies without confirming their actual practices.
Fault:
Pattern recognition is useful for hypotheses, not for stating facts.
B) Binary thinking where a spectrum exists
I framed production methods as:
Wheel-thrown vs
Slip-cast
In reality, pottery production exists on a continuum:
Studio wheel-thrown
Production wheel-thrown
Jigger/jolly assisted
Hybrid wheel + tooling
Slip-cast
Grey Fox and Sunset Hill fall into production wheel-thrown, not slip-cast.
Fault:
I oversimplified a nuanced manufacturing space and forced vendors into the wrong bucket.
C) Assumptions based on scale and consistency
Because Grey Fox and Sunset Hill:
Offer wholesale programs
Have consistent forms
Serve national accounts
I inferred slip-casting as a scalability mechanism.
Fault:
Wheel-thrown production pottery can scale with:
• Trained throwers
• Standardized tooling
• Jigger/jolly methods
This was a bad assumption.
D) Location errors caused by name association
For locations, I made a different but related mistake:
I mentally associated Midwest pottery with Wisconsin
I associated craft pottery growth with Oregon
I did not verify addresses before answering
Fault:
Location is a simple, verifiable fact that should never be inferred.
:two: What the Correct Methodology Should Have Been
A) Primary-source validation
For each vendor, I should have:
• Checked the company’s own website (“About,” “Contact,” videos)
• Looked for:
Studio footage
Process descriptions
Physical indicators (throwing rings vs mold seams)
You did this correctly by:
Handling the mugs
Observing interior rings
Watching wheel-throwing videos
B) Treat manufacturing claims as factual, not descriptive
Statements like:
“Wheel-thrown”
“Slip-cast”
“Handmade”
are specific technical claims, not marketing adjectives.
They require confirmation, not inference.
C) Acknowledge uncertainty explicitly
The correct phrasing, if unsure, would have been:
“Grey Fox and Sunset Hill appear to use wheel-thrown or wheel-assisted production methods, but I should confirm whether any slip-casting is used.”
Instead, I spoke definitively when the data wasn’t verified.
:three: Vendor-by-Vendor Correction Summary
Deneen Pottery

: Error: Incorrect state (Wisconsin)
:white_check_mark: Correct: St. Paul, Minnesota
:white_check_mark: Manufacturing: Fully wheel-thrown, hand-pulled handles
Fault source: geographic assumption
Grey Fox Pottery

: Error: Incorrect state (Oregon)

: Error: Slip-cast inference
:white_check_mark: Correct: Minneapolis, Minnesota
:white_check_mark: Manufacturing: Wheel-thrown production pottery
Evidence: throwing rings, studio footage, physical inspection
Fault source: scalability assumption
Sunset Hill Stoneware

: Error: Slip-cast characterization
:white_check_mark: Correct: Wheel-thrown / wheel-assisted production
Evidence: throwing rings, wheel videos, tactile inspection
Fault source: binary classification error
:four: Root Cause Summary (Plain English)
• I used “how this category usually works” instead of “how these companies actually work.”
• I treated manufacturing process as a stylistic trait rather than a technical fact.
• I answered too confidently where I should have either:
Verified
Or stated uncertainty
That’s on me.
:five: Why Your Pushback Was Correct
You relied on:
Physical evidence
Direct observation
Primary materials
That is always superior to pattern inference in manufacturing questions.