So, I'll say this with the caveat that I'm a writer who is troubled by ChatGPT and the fact that my writing (along with so, so, so many other writers) was used to help train the system without my consent, so I'm not a fan generally of generative AI. (Old lady shakes her fist at the sky)
I think the responses you got were surprisingly sophisticated and I can see how they'd feel helpful, but do you have any concerns about giving that level of personal information (and personal information about your partners, who didn't actually consent) to OpenAI, which will save it forever and use it for whatever purposes it deems acceptable?
Yeah, fair question! I did not give ChatGPT anything but first names. I also feel like so many companies have access to so much of our data right now; I'm guessing that the social media companies know so much more about me than what I shared with ChatGPT. Not that this means we shouldn't care or should be complacent, of course. This is food for thought, for sure, and I will keep reflecting on it!
I think of environmental impacts of AI and refrain from using it as much as possible. My search engine (none of us should be using google they’ve admitted to using it for surveillance) is turned off as to not allow automatic AI generated results. I agree with comments saying this is bleak, takes away from human connection and talking to friends for alternative perspectives, security risks, and how AI has stolen writing samples to improve its answers.
I like your substack a lot generally, but this article felt very strange. I share the concerns about privacy another reader voiced, and I also just don't enjoy reading paragraphs of terrible prose. It's depressing that your date had to turn to chat gpt instead of a group chat to brainstorm dating. I don't know, this is a poorly written comment, and I don't want you to feel demeaned. I just find this all very bleak and solipsistic and not at all "cuddly"/"validating"
I’m really bummed that people responded here and fixated on their own value judgements of chatGPT rather than the surprising insight that the dominant narrative now knows better than many men about what people need. I agree with your finding, Loretta. And encourage you to keep writing! I miss your updates and I think the journey you’re on is one many people feel too scared to confront. One person using AI twice as an experiment is not the ethical issue people are making it seem. This argument can be applied to so many things that self-police women’s sexuality.
I did the same thing recently out of curiosity after reading a comment on Reddit about ChatGPT making a good therapist. I thought there was no way I’d get anything out of it, but I ended up uploading the screenshots documenting the end of a very important friendship, to see if it could help analyze what I could have done better or differently throughout the conversation…and it was shockingly incredibly insightful and helpful and has done more to help me process the situation and dynamic than hours of therapy ever did.
There’s obviously something horrible and wrong with stealing other people’s work, and, at the same time, if an LLM can process everything we as humans have ever written down and somehow synthesize that in a way that helps me process the end of a very meaningful friendship? I guess I’m practicing holding two truths, as they say.
So, I'll say this with the caveat that I'm a writer who is troubled by ChatGPT and the fact that my writing (along with so, so, so many other writers) was used to help train the system without my consent, so I'm not a fan generally of generative AI. (Old lady shakes her fist at the sky)
I think the responses you got were surprisingly sophisticated and I can see how they'd feel helpful, but do you have any concerns about giving that level of personal information (and personal information about your partners, who didn't actually consent) to OpenAI, which will save it forever and use it for whatever purposes it deems acceptable?
Yeah, fair question! I did not give ChatGPT anything but first names. I also feel like so many companies have access to so much of our data right now; I'm guessing that the social media companies know so much more about me than what I shared with ChatGPT. Not that this means we shouldn't care or should be complacent, of course. This is food for thought, for sure, and I will keep reflecting on it!
I think of environmental impacts of AI and refrain from using it as much as possible. My search engine (none of us should be using google they’ve admitted to using it for surveillance) is turned off as to not allow automatic AI generated results. I agree with comments saying this is bleak, takes away from human connection and talking to friends for alternative perspectives, security risks, and how AI has stolen writing samples to improve its answers.
100% agree with this comment. AI is trained on other people’s writing/art/IP, which is stolen from them, and is terrible for the environment.
I like your substack a lot generally, but this article felt very strange. I share the concerns about privacy another reader voiced, and I also just don't enjoy reading paragraphs of terrible prose. It's depressing that your date had to turn to chat gpt instead of a group chat to brainstorm dating. I don't know, this is a poorly written comment, and I don't want you to feel demeaned. I just find this all very bleak and solipsistic and not at all "cuddly"/"validating"
This is all fair! I appreciate you sharing your feedback!
I’m now picturing ChatGPT as a big cuddly Baymax (Your Personal Healthcare Companion).
It actually IS kinda cuddly! It's so empathetic, encouraging and validating!
I’m really bummed that people responded here and fixated on their own value judgements of chatGPT rather than the surprising insight that the dominant narrative now knows better than many men about what people need. I agree with your finding, Loretta. And encourage you to keep writing! I miss your updates and I think the journey you’re on is one many people feel too scared to confront. One person using AI twice as an experiment is not the ethical issue people are making it seem. This argument can be applied to so many things that self-police women’s sexuality.
I did the same thing recently out of curiosity after reading a comment on Reddit about ChatGPT making a good therapist. I thought there was no way I’d get anything out of it, but I ended up uploading the screenshots documenting the end of a very important friendship, to see if it could help analyze what I could have done better or differently throughout the conversation…and it was shockingly incredibly insightful and helpful and has done more to help me process the situation and dynamic than hours of therapy ever did.
There’s obviously something horrible and wrong with stealing other people’s work, and, at the same time, if an LLM can process everything we as humans have ever written down and somehow synthesize that in a way that helps me process the end of a very meaningful friendship? I guess I’m practicing holding two truths, as they say.
IME it’s pretty good too! but at the end of the day i find i still have to trust my gut(the part AI doesn’t have)
it’s so helpful at organizing thoughts into points