A few months ago, I was sick with the flu and ended up writing a book, which I really enjoy as a hobby. To get some instant feedback, I started sharing my chapters with ChatGPT for proofreading. Although I had some privacy concerns at the time, I can't quite remember what I did regarding settings. Recently, I was looking into some of the Meta drama and noticed that ChatGPT mentioned needing to enable the training model function explicitly. When I checked, I found that it was already turned on, which I definitely didn't set myself. Has there been a feature update that changed how these settings are named? I found a lot of features online that don't seem to match what's available now. I've turned off the training model option, deleted my chats, and even emailed support to ask them to wipe my data. I'm just looking for some reassurance that my work won't be used or leaked in any way. Is it really just used to train language models? I know I should have been more careful originally, and I'm feeling pretty anxious about this whole situation.
3 Answers
I can relate to your worry! I think the risk comes if you’re dealing with something niche. If your writing isn’t wildly unique or specialized, you should be fine. Just keep monitoring your work and protect your original drafts!
It's normal to feel anxious about sharing original work online. I’ve had similar fears, but most of the time, unless you're writing really specialized content, it’s rare for your exact words to turn up. A lot of it gets mixed and matched in unrecognizable ways. Just be cautious and keep backups!
Honestly, I wouldn't worry too much. Your input is just one drop in an ocean of data. The AI uses lots of different examples to generate responses, so your work is unlikely to be recognized or reproduced exactly. It’s more about patterns than specifics.
That’s reassuring! Thanks for the insight!
I completely understand! Thanks for your reassurance!