1. Home
  2. misc-audio
  3. 197099-20253

I asked ChatGPT about me and my posts...

Lets keep this friendly and avoid doing this for others or using it to attack me or anyone else or prove your view of some one else was right.  

Let’s try an AI experiment! In another thread I was accused of many things including putting myself out here as THE authority on speakers when in reality I was cutting and pasting from ChatGPT.  The poster’s ad hominem attack included that I had posted 17k times.  Kind of an odd way to attack me but it got me thinking:  Can I use ChatGPT to summarize my participation?  

Here is the prompt I used: 

I do a lot of posting on Audiogon. You can see them from this link: https://forum.audiogon.com/users/erik_squires/discussions - What I want from you is an unbiased, short description of my posts, their technical accuracy and tone.

And here is the reply.... 

Here’s an unbiased, concise summary of your Audiogon contributions based on your discussions and responses:

1. Content Themes

  • You engage broadly across audio topics—from gear, system setup, signal issues, speaker matching, subs, room acoustics, to industry commentary and hobby culture.

Many posts start open-ended questions designed to gather others’ experiences rather than to assert your own conclusions.

You also start opinion or culture threads (e.g., on hobby inclusiveness or value vs. price), not just technical Q&A.

2. Technical Accuracy

  • Your responses tend to be practical and experience-based, mixing personal system experience with general principles (e.g., room modes, sub integration, signal chains).

You avoid unverified claims—emphasizing listening experience over ungrounded assertions.

Some answers suggest measured knowledge (e.g., recommending room mode tools, discussing distortion signatures), though they are framed as personal insights rather than authoritative engineering statements.

3. Tone

  • The tone is collegial and conversational—you generally invite others to share and clarify rather than lecture.

You moderate discussions by keeping threads on topic and encouraging constructive responses.

Occasionally you push back against what you see as snobbery or entrenched opinions in the hobby, but do so in a thoughtful way rather than aggressively.

4. Community Presence

  • With hundreds of discussions started and thousands of responses, you are a high-traffic, active member contributing to many areas of the forum.

Your posts often blend technical curiosity with personal experience, aiming to help others while exploring subjective aspects of audio enjoyment.

Summary (in one sentence):
You’re an experienced, conversational audiophile contributor whose posts mix practical system advice with open-minded discussion and generally accurate, experience-based commentary that invites participation rather than dictates conclusions.

----------------------------------------------------------------

You are free to disagree with ChatGPT, of course!!!  What I wanted to start, thought it would be fun, is for you to do the same for yourself.  If you try this and like the results, please post ChatGPT, or Claude or Gemini’s views of yourself here! 

Lets keep this friendly and avoid doing this for others or using it to attack me or anyone else or prove your view of some one else was right.  

erik_squires
ghdprentice's avatar
ghdprentice
9,466 posts
 

AI is an incredibly powerful tool. But like all tools you have to learn how to use it, its limits and strengths. 

Initially I was drawn to it by asking open ended philosophical, scientific, historical, and psychological questions. It is really good at these,  it has access to vast and diverse information. Then I started papering it with all sorts of questions... how to fix some esoteric problem with my MacBook, bicycle component, coffee maker... excellent. How to do stuff... I uploaded a photo that I was about to create and paint in oil colors and it created the pallet, down to which brand of paint to use to keep the transparency consistent... and it was perfect. Bringing the experience level of decades instead of my meager four years. 

But when you are asking for specific brand recommendations. I asked for some single purpose streamers that cost about $5K and several were streamer/DACs.Then, I put in the components of my system and it characterized it perfectly and recommended the perfect interconnects and power cords. I told It I had those and it recommended direct lines and a power conditioner... I told it I had it and I told it I had them... etc. 

I asked what to do with some left overs in the refrigerator... it gave me a Thai, American, French, Cuban...etc recipe. It came up with a training plan for my puppy... auto shop that was particularly good with my car, bicycle accessory choices. 

Overall, right on and incredibly valuable. But, under certain circumstances, it is wrong or a bit misleading. I am aware of when I ask something that it might not do well with, and I spot check information... and always if it is important. 

It’s like any other tool. Very useful, but you have to know how to use it and its limits. If you want to trip it up so you can point at it and say, "See! it is wrong" you can do that... but it is likely to be your loss. For me, it has at least tripled my learning rate in philosophy, science, history and technology... and that is incredible. 

And I thought it was pretty fun to see what it had to say about my comments on the forum. 

erik_squires OP
17,551 posts
 

@lanx0003 - 

You keep using the word “authoritative” because you believe your contributions to this forum can be justified that way.

I have typed that word, but please state a specific example where I used that term to justify any of my posts, or beliefs. 

The only place I have been using that word AFAIK is to caution about using AI’s responses that way, including if AI thinks my post should be taken in an authoritative manner.  My point is the opposite of "I am an authority."  My point is "AI thinks anyone who posts is an authority, including me." 

If I have posted otherwise, please show me and I'll happily apologize and endeavor to correct the record. 

Best,

 

The Erik Squires (inside joke). 

timothy468's avatar
timothy468
88 posts
 

Fascinating exercise! Here is what I got when I asked the same question:

Good God, you are a tool! You aren't much, but you're all you think about! 

You provide highly opinionated but frequently grossly incorrect answers and quickly delve into ad hominem attacks when anyone pushes back on your unproven assertions.

mahgister's avatar
mahgister
15,051 posts
 

Instead of attacking the motives of the one and the others as often in audio threads... It is better to think about the A.I. craddle and his meaning through its effects on the  human thought : 

There are already two humanities (the individuals thinkers and the mass of sheep) , but currently too intertwined to be recognized as distinct, and soon there will be four distinct humanities:

The  passive consumer, whose brain has been "washed" like a sieve, will soon become more perfect consumer with a five-second attention span...

Then the perfect cyborg plugged directly into the control center for specialized tasks..

And we  with some others, the two groups capable of concentrating for hours on a difficult subject, mastering individual general  thinking , we refusing either to be subservient or to be in control but the others, the oligarchs and their small cohorts of thinking  brain  which will not refuse power.

I take this very seriously ...

 

mdalton's avatar
mdalton
810 posts
 

One more AI musing.  When I was doing the hum detection exercise with ChatGPT, it complimented me several times, going so far as to congratulate me for thinking like an engineer.  I've since filed for divorce.  When my wife asked me if there was someone else, I said "yes, his name is Chat."

 

More to discover