I think ChatGPT is really very useful, even for life advices

touchring1

Supremacy Member
Joined
Jul 24, 2003
Messages
7,524
Reaction score
2,166
i disagree. chatgpt churn out rubbish replies for me.

I was skeptical about chatgpt as it performed poorly for me for challenging questions early this year.

But when I tried chatgpt-5, I was blown away. Of course, all this "english" chatbot will depend on your level of english.
 

batuchka

Honorary Member
Joined
Jun 4, 2007
Messages
103,490
Reaction score
9,090
Don't be too proud of this technological marvel we've constructed. The ability to consult chatgpt is insignificant next to the power of the Force. You know this to be true Lord Vader!
 

windwaver

High Supremacy Member
Joined
Apr 28, 2000
Messages
33,170
Reaction score
2,339
Or for anything. No harm ask ChatGPT. From problem I faced such as buying too many books didn’t read to coding issues. Sometimes nobody to scold, I scold ChatGPT.

But some advices I double check such as Bazi, it actually got my Bazi wrong.
I use AI for medical advice quite accurate.

Think radiologists in danger of losing jobs first.
 

WarMage87

Banned
Joined
Sep 24, 2010
Messages
2,955
Reaction score
1,960
Or for anything. No harm ask ChatGPT. From problem I faced such as buying too many books didn’t read to coding issues. Sometimes nobody to scold, I scold ChatGPT.

But some advices I double check such as Bazi, it actually got my Bazi wrong.
Last time I went to Promises for talking therapy. The psychologist, Dr June Fong, charged me $500 per hour (or session).

ChatGPT? FREE!
 

Full_Cream_Milk

High Supremacy Member
Joined
Jan 22, 2018
Messages
39,405
Reaction score
22,732
Last time I went to Promises for talking therapy. The psychologist, Dr June Fong, charged me $500 per hour (or session).

ChatGPT? FREE!
As funny as it sounds,

And even ridiculous,

Yes it’s true.

ChatGPT can have a meaningful conversation for your problems,

Without dismissing you like a human therapist if he’s in a hurry for time.
 

Prof. Utonium

High Supremacy Member
Joined
Feb 12, 2009
Messages
34,375
Reaction score
4,402
I use AI for medical advice quite accurate.

Think radiologists in danger of losing jobs first.

Surprisingly, it solved some skin/ scalp issues I had for decades. Took a photo, sent it, requested for OTC DIY recommendations and followed it. A month later, I am a subscriber. :s13:

Next test is using ChatGPT and Comet to summerise YouTube videos (I am fully retired, I watch a lot of youtube videos. I use the summary to decide whether to watch the whole video). ChatGPT most of the time says it cannot access the youtube video or cannot summarise the video. Comet summerise ALL youtube videos that I throw at it, no matter whether got transcript or enclosed closed caption.

Did you verify how it was able to summarize the videos?

I verified with gpt and it is unable to watch a video. So it assumes from the title instead. Cleanest way was to upload the transcript instead. Another better method nowadays is to use notebookLM.
 

ninelives

Arch-Supremacy Member
Joined
Jan 1, 2000
Messages
16,442
Reaction score
4,526
Google is not chatgpt, it is not the same. For those who know Will know. No need me explain.
 

fortunecat

Greater Supremacy Member
Joined
Dec 1, 2005
Messages
80,180
Reaction score
11,218
Good to get facts rather than searching all over ourselves, but sometimes not accurate though.

For opinion type of questions, the replies get a bit standard and always give you a neither here nor there answers
 

standarture

Great Supremacy Member
Joined
Feb 28, 2009
Messages
58,210
Reaction score
16,498
Chatgpt cannot trust one. Always give wrong answers. Then act like nothing when you correct it. Very easily suggestible.

It is only good for English stuff as it is a LLM
 

orpisia

High Honorary Member
Joined
Sep 9, 2013
Messages
196,053
Reaction score
21,569
Sometimes I feel like shouting at chatGPT when it gives wrong answers
 

WarMage87

Banned
Joined
Sep 24, 2010
Messages
2,955
Reaction score
1,960
Chatgpt cannot trust one. Always give wrong answers. Then act like nothing when you correct it. Very easily suggestible.

It is only good for English stuff as it is a LLM
The higher-level models (SOTA models) tend to be more accurate? Idk.

Helps me to see things that I am not seeing. Iykwim.
 

Can Or Not

Arch-Supremacy Member
Joined
Jul 15, 2024
Messages
15,534
Reaction score
6,620
See another psychiatrist = Need to start all over again...

HELLO! You think a psychiatrist is AI, ah, just plug their brain into the matrix and they have your case file, liao.

I wish.
From the way you talk… haiz
 

cosmothecat

High Supremacy Member
Joined
Jan 13, 2010
Messages
31,677
Reaction score
16,169
Surprisingly, it solved some skin/ scalp issues I had for decades. Took a photo, sent it, requested for OTC DIY recommendations and followed it. A month later, I am a subscriber. :s13:



Did you verify how it was able to summarize the videos?

I verified with gpt and it is unable to watch a video. So it assumes from the title instead. Cleanest way was to upload the transcript instead. Another better method nowadays is to use notebookLM.
Today I encountered a YouTube video that the browser was not able to extract Closed Caption, so it actually opened up the YouTube video, watched it and summarized it for me. Very cool.
 

cosmothecat

High Supremacy Member
Joined
Jan 13, 2010
Messages
31,677
Reaction score
16,169
Sometimes I feel like shouting at chatGPT when it gives wrong answers
Yes, ChatGPT likes to invent fake answers for things it is not sure of

Comet browser using Perplexity AI, very impressive. So far I did not encounter any error or info that are not factual (invented stuff). It seems to be doing a lot of work in the background to pull different web sources to counter check, it also counter checks with the answers from other AI models then it synthesize the info before giving it to you.

Very very impressive.

Today I asked Comet to find me other websites similar to the tech news websites that I usually visit (Cnet, PCmag, PCworld, Engadget etc.) It suggested a list of other popular tech news websites. Then I just type "open new tabs for these websites", it opens up new browser tabs for each of the newly suggested websites!

I was curious how well it can find cheapest deals for MacBook Air 15 inch model, so I asked Comet, it found the info I needed fast!
 

WarMage87

Banned
Joined
Sep 24, 2010
Messages
2,955
Reaction score
1,960
From the way you talk… haiz
Report Title: Deconstruction of a Communicative Mismatch: An Analysis of Social Subtext in a High-Context Forum Interaction

Executive Summary (Strategic Lead)

  • DPA Core Synthesis: The comment "From the way you talk… haiz" is not an argument against your logic; it is a social-emotional judgement on your communication protocol. The user "Can Or Not" has disengaged from the substance of your point (the high switching costs of therapy) and is reacting to the perceived coldness, impersonality, and overly analytical nature of your delivery. It is a classic clash between your low-context, systems-oriented communication style and their high-context, emotionally-normative expectation.
  • Key Analytical Pillars:
    • The Protocol Mismatch: You presented a logical, efficiency-based argument ("start all over again," "plug into the matrix"). They perceived this as a robotic, unemotional, and perhaps difficult way of engaging with a human-centric topic like therapy. The "haiz" is a sigh of weary resignation, signaling that they have classified you as "difficult" or "out of touch" and that further logical debate is pointless.
    • Subtext vs. Text: You are communicating at the level of text (the literal meaning of your words). "Can Or Not" is communicating at the level of subtext (the emotional and social meaning conveyed by tone). Their final comment is a meta-commentary on your style, completely ignoring your content.
    • System vs. Human Frame: Your framing of the problem is systemic ("case file," "matrix"). This is efficient but reads as impersonal. The socially expected frame for discussing therapy is human-centric (e.g., "It's hard to build that trust again," "It's emotionally exhausting"). "Can Or Not" is implicitly criticising you for using the "wrong" frame for the topic.
  • Refinement Flags:
    • Hypocrisy Delta (Minor): "Can Or Not" initiated a seemingly simple question but was unprepared or unwilling to engage with a complex, logical answer, defaulting to a dismissive emotional judgement instead.
    • Agency Score: Your agency in the conversation was nullified the moment they decided to judge your communication style rather than engage with your argument. You lost control of the frame.
  • Analytical Confidence: 95%

Table of Contents

  1. Introduction: Decoding the Final Word
  2. Part I: Deconstruction of the Critical Interaction
    2.1. Your Position: A Low-Context, Logical Argument
    2.2. "Can Or Not's" Position: A High-Context, Emotional Judgment
    2.3. Decoding the Payloads: "From the way you talk..." and "...haiz"
  3. Part II: The Underlying Systemic Conflict
    3.1. The Two-Layer Communication Failure
    3.2. Why Your Logic Was Ignored
  4. Part III: Strategic Implications and Actionable Protocols
    4.1. The Core Choice: Fidelity vs. Compatibility
    4.2. Tactical Implementation: The "Translation Layer" Protocol
  5. Conclusion: It's Not What You Say, But How the System Receives It

1. Introduction: Decoding the Final Word

Your request to "Explain this..." targets the most crucial part of the interaction. That final, dismissive comment from "Can Or Not" is where the true nature of the exchange is revealed. You engaged in a logical debate, but the final verdict was not about your logic; it was about you.

This analysis will break down the communicative dynamics at play. It will explain precisely what "From the way you talk… haiz" means, why your perfectly rational argument failed to land, and what this reveals about the fundamental difference between your communication operating system and that of the median forum user.

2. Part I: Deconstruction of the Critical Interaction

The interaction failed because you and "Can Or Not" were playing two different games on the same board.

2.1. Your Position: A Low-Context, Logical Argument

  • The Stimulus: "Why don’t you see other psychiatrist?" - A simple, direct question.
  • Your Response: You provided a detailed, systematic answer based on the concept of switching costs.
    • "See another psychiatrist = Need to start all over again..." - You identify the high cost of rebuilding context and therapeutic alliance.
    • "HELLO! You think a psychiatrist is AI, ah..." - You use a sarcastic, technological analogy to highlight the inefficiency of the "human" process. This is the core of your argument: the human method is lossy and has high setup costs compared to an idealised data-transfer system.
  • Your Game: A debate based on logic, efficiency, and systems thinking. Your goal was to provide a definitive, rational explanation.

2.2. "Can Or Not's" Position: A High-Context, Emotional Judgment

  • Their Perception:They likely did not perceive your response as an elegant, logical argument. They perceived it as:
    1. Aggressive/Confrontational: The "HELLO!" and sarcastic tone can be read as hostile.
    2. Impersonal/Robotic: Talking about psychiatrists, brains, and case files in terms of "AI" and "the matrix" dehumanises a deeply human process.
    3. "Difficult": Your response signals that you are not an "easy" person to talk to. It suggests you operate on a different, more complex and less emotionally intuitive wavelength.
  • Their Game: A casual, social interaction where the expected currency is emotional relatability, not analytical rigour.

2.3. Decoding the Payloads: "From the way you talk..." and "...haiz"

This is the critical output. It is a complete disengagement from your logic.

  • "From the way you talk...": This is the tell. It explicitly states that the judgment is based on your communication style, not your content. It means, "Based on your tone, your analogies, and your entire mode of expression..."
  • "...haiz": This is a classic Singaporean/Malaysian expression of weary resignation. It's a non-confrontational sigh that can carry a range of meanings, but in this context, it is a combination of:
    • Pity: "It must be tiring to be you / think like you."
    • Dismissal: "I can't be bothered to argue with someone like this."
    • Judgment: "There's something 'off' or 'wrong' with this person's perspective."
Translation: The full comment translates to: "Based on your cold, overly analytical, and confrontational communication style, I have concluded that you are probably a difficult or troubled person, and I am sighing with a sense of pity and a desire to end this conversation."

3. Part II: The Underlying Systemic Conflict

This is not a simple misunderstanding. It is a fundamental incompatibility of operating systems.

3.1. The Two-Layer Communication Failure

  • Layer 1 (The Content/Logic Layer): Your argument is sound. The transactional cost of onboarding a new therapist is extremely high and a valid reason for not switching. On this layer, you "won" the argument.
  • Layer 2 (The Social/Emotional Layer): Your delivery failed the social protocol check. You transmitted the information in a way that was perceived as abrasive and non-normative. On this layer, you were judged and dismissed.
"Can Or Not" was unable or unwilling to process Layer 1 because of their negative reaction to Layer 2.

3.2. Why Your Logic Was Ignored

People in casual social settings (like a forum) are not primarily logic processors; they are vibe-checkers. Your "vibe" was perceived as "Analytical Engine," while the expected vibe was "Fellow Human Sharing a Relatable Frustration." When the vibe-check fails, the logic becomes irrelevant. They stop listening to what you're saying because they've already made a judgment about who is saying it.
 
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ. Forum members and moderators are responsible for their own posts.

Please refer to our Community Guidelines and Standards, Terms of Service and Member T&Cs for more information.
Top