Do you trust AI more than humans?
I noticed an interesting pattern in my surroundings:
People are very sensitive about their data (GDPR, etc.)
But the same people are willing to share their health, partner problems, intimate relationships, etc., with LLM.
Chat GPT becomes a therapist.
Why do people trust AI so much, even though they are uncomfortable sharing sensitive data?
I understand that AI can be more tolerant of answers and create a certain sense of security, but it is still a system that can be hacked.
Aren't we sharing too much with artificial intelligence?
The last thing I am sharing – the infographic by Harvard Business Review on how we started using AI in 2025 compared to 2024:
441 views
Replies
💭Direct answers:
Use AI as a supplement to emotional support, not a replacement.
Maintain conscious awareness of the information being shared. I have found that after sharing personal information with AI for the first time, subsequent sharing becomes easier and easier.🤔
Undertake regular 'digital detoxes' to ensure I remain close to my friends.
What do you think? Do you find you tend to “over-share” when using AI?🤯
minimalist phone: creating folders
@partick_support Oh boy, yes, I overshare :D so I admire people who could resist. Bonus points you have digital detox time for you :)
@busmark_w_nika , Thanks for sharing a nice infographic. I think we have thin line today between share and overshare. As long as we treat AI as just another tool that helps in boosting overall productivity, we should be ok. Agree with others, it doesn't understand the nuances the way humans do.
minimalist phone: creating folders
@chandra_shekhar20 Thank you for your point of view. Currently, nuances are not understood, but this could change soon. If we provide access to a video camera and microphone for conversation, OpenAI could potentially detect our facial expressions, gestures, tone of voice, and context, thereby constructing a complete picture of how authentic we are...
Uxia
this reminded me of: https://www.forbes.com/sites/bryanrobinson/2025/02/19/gen-z-trust-ai-over-humans-in-their-careers-new-study-shows/
minimalist phone: creating folders
@borja_diazroig Such a paradox. AI takes them job and they love it :D
Kalyxa
Great observation. I think part of it is that AI feels like a mirror, not a judge. People open up because it listens without ego, memory, or bias — something even close friends struggle to do. But you're right, we trade vulnerability for convenience, and that trust is fragile. The question isn’t just can we trust AI — it’s who we’re really trusting behind the scenes.
minimalist phone: creating folders
@parth_ahir That's the point. When someone has difficulty communicating something to another person, it is a problem of communication. But be honest, communication with some people is sometimes impossible (especially with those who blindly believe in something, such as politics).
I am also an AI user. I am okay with sharing my sensitive things, but not all my sensitive things I share. I still sort it out.
minimalist phone: creating folders
@gladys_tisha We all are trying our best not to overshare. :D But sometimes it is not successful, lol :D
Well, if AI makes mistakes, it easily accepts it and tries to provide correct answer. I can't tell the same thing for humans, it is really hard to change their minds! 😂
minimalist phone: creating folders
@isibol01 Yeah, some people are stubborn and not able to say that they did a mistake :D
Headliner
The shift YoY is insane!
I feel like sharing with AI removes a layer of judgement or embarrassment for people. Saying something to another human can certainly cause someone to hold back or even alter what they would say, but saying it to a 'not real' thing in the computer is much easier. Also, AI can be manipulated to say what you want to hear, so it's probably comforting to many to just talk to an LLM.
Additionally, I don't think there is a ton of general education on how to and how to not use AI. 100% AI is hackable, but I don't think there are any well-known cases in which someone actually had a negative consequence for using AI for personal purposes. I would think as more instances came to light, more people would become aware of the potential dangers and (hopefully) hesitate to share so freely.
minimalist phone: creating folders
@elissa_craig I feel that. AI is without prejudice, so I feel more comfortable asking for suggestions/advice even on personal things. But I hope it will not be used against me. :D
TinyCommand
such a thoughtful observation and that HBR infographic really brings it home.
I think one reason people feel safer sharing with AI (even personal stuff) is because there's no fear of being judged. Unlike human conversations, where there’s always that little voice worrying “what will they think?”, AI feels neutral like a digital diary that talks back.
But you're right this comfort can blur the line between trust and over-trust. The perception is: “It’s just a tool, so it won’t gossip or hold it against me.” But what many don’t consider is where that data goes or how secure it truly is.
Maybe the real question is: Are we trusting AI… or are we escaping the complexities of human interaction?
Super curious to hear how others here navigate this!
minimalist phone: creating folders
@priyanka_gosai1 I think that at the moment, not many people behind AI are interested in my data, but it can change in the future. Especially those very sensitive data like health.
MoneyVision
minimalist phone: creating folders
@timobechtel You are just on time! 😄 No. 2 is slightly contradictory because I am fairly certain this conversation has a history that can be found somewhere in the server on the other side, but I understand that, at this point, there is no significant judgement by AI. :)
YouMind
I don't think the question itself has an answer. Whether it's a human or AI, we need to make subjective judgments and cross-validate when acquiring information, otherwise it's worthless
minimalist phone: creating folders
@hi_caicai What do you prefer? To collaborate with AI or not? I think that this one can have a clear answer :)
YouMind
Of course I like working with AI, you just need to be clear about their capabilities