pull down to refresh

If a smart friend tells you something that ChatGPT or some other AI tool disagrees with, who do you instinctively trust more?
Let’s assume you don’t have the time to dig into the issue and probe for follow up questions… who do you trust more in the moment?
Let’s also assume the smart friend is talking about a topic which you consider them to be knowledgeable.
Wow, basically nobody picked AI.
Curious to see if this changes over the coming years...
reply
I would trust my friend but I would dig into it later.
reply
I'm getting mental orgasms when people around me know some real world information without internet and can keep conversation without smartphone
reply
The smartness of the friend is less important than the type of person the friend is. You don't have to be smart to be correct, so if it's the type of person who is smart enough to only say ehat they know to be true, then the person.
reply
By design, LLMs don't know if they are right or wrong. My friends usually do. So the answer to your question seems obvious to me.
reply
well said
reply
Friends for sure. Thing is, it's not hard to confirm things either way.
reply
AIs are wrong pretty often. I call it out and it doesn't get pissed. That's a plus
reply
The biggest mistake I see in younger folk that grew up with the Internet is their assumption everything is on the web though. I call It the Wikipedia effect.
reply
Chat bots are designed to entertain and satisfy us... not find the truth. Humans are designed to find truth, even when it upsets us.
reply
I think trust isn't binary. AI is like a library—fast, vast, but soulless. A friend brings something AI can’t replicate: wisdom shaped by life, intuition, and context. In the end, if it's someone I truly respect, I trust the heart and experience over the machine. But the real answer is... I stay skeptical of both.
reply
Don't trust, verify.
reply
This is interesting because I recently had a conversation with a friend, who told me that chatgpt was wrong about a small detail on a very niche topic he happens to be very knowledgeable about. I trust he's right in this case, because I am well aware of AI "hallucinations"
In another instance, I was researching a specific topic with the help of ai, and got some information contradicting the premise of a book I had recently read. I made sure to double check other sources, and yes, chatgpt was right, and the book was wrong ! Can you imagine writing an Amazon 5-star rated book, and the premise of your argument is factually and demonstrably wrong, and most readers won't even know?
reply
I can trust the AI. Humans are unpredictable and tend to get emotionally involved most of the times. My opinion is that as long as you know how to handle the AI, it can be really great and precise.
reply
reply
In my opinion, the response of an intelligent, educated, and trained person is more reliable than a chatbot. Artificial intelligence still makes mistakes when answering very specific questions. The human brain is still superior in its functioning.
reply
I don’t ask ChatGPT
reply
Friend > AI
reply
No one is fully trustable so both are may be necesary nowadays
reply
I don't blindly trust either. Whether it's a smart friend or an AI, verification is mandatory.
reply
I would assume it's a question with a lot of nuance where how you answer it could depend on certain assumptions or contexts which may be unspoken
reply
Let’s also assume the smart friend is talking about a topic which you consider them to be knowledgeable.
I would trust my friend more.
reply
I would trust my friend more, but I would also use chatgpt, I suppose, to corroborate the information.
reply
I prefer smart friend.
Because if he/she were wrong I can ask the possible failure reasons. 😄 You can't learn from machine.
reply