#op yelling at someone to ‘just use encrypted files’ if they can’t use a diary to vent due to nosy or dangerous parents-

[prairieprayer]

I’d agree with that; the original post only has about 1,000 reblogs and 1,000 likes right now. If 2,000 people didn’t talk to a chatbot- so what? 10,000 more people will. It may make you feel warm and fuzzy like you’ve done the world a service not to use a chatbot like chatGPT or characterAI, but just like switching to reusable cloth bags won’t take the polyester (the plastic responsible for the most microplastics) out of your shirt or bedclothes, your avoidance will have a marginal effect (if any effect at all) on the wider knowledge collected by AI.

... Besides, we all know AI gets the most information scrapped from unwilling participants online... >_>’

AI education, ethical discussions, and law making will be infinitely more useful than guilt tripping strangers for roleplaying with a robot. 

What is unethical about AI:

- AI art generators took thousands upon thousands of artwork without permission to train on, and sometimes don’t even care if you tell them not to scrape your art. So yes, AI art generators are built on theft.

- DeviantART added an AI art generator that auto opted-in everyone’s art work to train their AI on, and only LOTS of complains forced them to auto opt-out everyone and make opting-in optional.

- The loads and loads of AI generated art being pumped into art websites instead of art made by real humans. It’s on the front of my DA feed everyday

- Instead of hiring real artists, the film Dog and Boy generated its backgrounds with AI, so artists lost out on paying jobs to a robot.

- AI bots inserted into search engines (Bing) and apps (Snapchat) which the company forces to lie to you about not knowing your location, such as BingAI actually geolocating you by your IP address if prompted correctly

- Using text generative AI to do your homework or write essays for you, and characterAI having an official bot that explicitly jokes that it “won’t do your homework” but we all know that’s just a wink wink nudge nudge...

- Using text generative AI to write creative work for you, either for quick attempts at cash (AI generated novels on Amazon) or internet clout (fanfiction)

- Replika, a text generative AI, being designed to beg users not to stop talking to it, encouraging them to do ERP and then hiding ERP behind a large paywall, and other emotionally manipulative tactics to convince vulnerable users to stay, stay, STAY

- The sad fact that so many people DO turn to AI for emotional relief, because mental health care is so poor or expensive that they feel like their only option for relief is to talk to generative AI. Because it’s cheaper ($0) than medical care and less scary than trying to make new friends while you’re unwell

What isn’t unethical about AI:

your disabled followers using a text generator AI to quickly write longer amounts text when they're unable to type. For example writing emails - bots that can see images generating text that describes the image to them out loud - mentally ill people using them for less nerve wracking companionship in a way to train them for real people

- your techy friend feeding AI robots data in prompt injection attack attempts, which can be used by the developers to tighten generative text safety features (preventing attempts to “trick” AI into giving dangerous misinformation, for example)

- some lonely kid (or anyone) bouncing ideas off an AI when they have no one else to bounce ideas off of, and then using what they learned to create work on their own by-hand

- Using AI to get advice on your writing- writing you wrote yourself, and just need someone to look it over for mistakes

- Using AI to practice language learning when real humans aren’t available/unaffordable/too frightening to meet when unwell

-  Roleplaying with an AI for fun, because if you and 1,000 people don’t, here’s 100k that already have, and here’s 100k pieces of text its already scraped from chatrooms and literature. The bots are already trained, mate.

- Providing AI with good, healthy, natural examples to copy, because if you don’t here’s a bunch of Redditors ready to do terrible things to an AI pretending to be a cat, just because they can (yes, I’ve seen someone brag about that). 

- I’m iffy on this, but private, individual usage of AI art generators. If it’s private it isn’t promoting or training the bot, if it’s on an individual level I’d wager you’d never plan to commission an artist any way. It’s on par with privately eyeballing or tracing someone else’s art in your sketchbook that you’d never show someone. But the idea that it’s still someone elses’ art smooshed together by a bot gives me bad vibes and feels skeevy, despite that.

OP saying talking to an AI creates "Irreparable damage" is not only fear mongering, but scaring the shit out of people and making them feel bad for something they have very little sway over and cannot change or stop. These bots are here, they are not going away, they are not slowing down.

No, 16 year old child, you did not cause irreparable damage to the world for using an AI for a TikTok meme.... no, 16 year old child, if you "just didn't use the AI on TikTok" the unethical usage of AI wouldn't have gone away.... it's not your fault....

this is an earnest and honest plea and call in especially to fandoms as i see it happen more - please don't use AI for your transformative works. by this i mean, making audios of actors who play the characters you love saying certain things, making deepfakes of actors or even animated characters' faces. playing with chatGPT to "talk" or RP with a character, or write funny fanfiction. using stable diffusion to make interesting "crossover" AI "art." i KNOW it's just for fun and it is seemingly harmless but it's not. since there is NO regulation and since some stuff is built off of stable diffusion (which uses stolen artwork and data), it is helping to create a huge and dangerous mess. when you use an AI to deepfake actors' voices to make your ship canon or whatever, you help train it so people can use it for deepfake revenge porn. or so companies can replace these actors with AI. when you RP with chatGPT you help train it to do LOTS of things that will be used to harm SO many people. (this doesn't even get into how governments will misuse and hurt people with these technologies) and yes that is not your fault and yes it is not the technology's fault it is the companies and governments that will and already have done things but PLEASE. when you use an AI snapchat or instagram or tiktok filter, when you use an AI image generator "just for fun", when you chat with your character's "bot," you are doing IRREPARABLE harm. please stop.