3 ms·
Paraphrasing a bit, and pulling in other recent OpenAI mews. "We know AGI is risky, and we've stated that while also shipping new products every 6 months. We h
by _heimdall 2y ago
Paraphrasing a bit, and pulling in other recent OpenAI mews.
"We know AGI is risky, and we've stated that while also shipping new products every 6 months. We hope you are excited by the recent announcement that OpenAI is looking into how to allow our customers to responsibly generate erotica, and want you to know that we are doing this with full concern of the risks and potential outcomes. We look forward to offering multimodal interactions with models so life-like that you may not realize you are interacting with a computer at all. And rest assured, we've fully thought through and planned for the potential dangers of such a future."
- aaomidi 2y agoTf is the risk of generating erotica?
- kumarvvr 2y agoGeneraring porn with children in it.
- yanderekko 2y agoDeepfakes?
- aaomidi 2y agoHonestly I was under the impression erotica is text based only. I guess not.
- _heimdall 2y agoOh I can see that. Given OpenAI's push for multimodal engagement with GTP4o I fully expect the erotica they want to support would be more than just text. I do agree with you there though, I can't think of any meaningful risks of text-only erotica unless you consider the psychological impact of the person choosing to engage with it.
- _heimdall 2y agoChild pornography is the easy answer. Deep fake porn would be a huge risk in my opinion though, kids making fake porn and sending it around school could absolutely lead to suicides or attacks against other students. Better question, how the heck do you see something like the ability for anyone to make extremely convincing, fake erotica and not see at least some risk in that?
- francasso 2y agoThey are going to say the usual things, child pornography, risk of terrorist use etc... and conclude that we need our overlords to protect us. Same reasons that have been used in the past to take away freedoms. Not saying they aren't real risks... Are there going to be sad incidents where innocent people will be harmed? Yes. The question though is if there will be enough of them, and egregious enough, for us to disregard the upsides of technology that is not heavily regulated.
- _heimdall 2y agoOh that's so interesting, I fall very much on the side of avoiding regulations as the solution at all costs. Regulations won't fix a damn thing and regulators have no chance of keeping up even if they wanted to. We need these tools to not be developed or offered at all by choice. Unfortunately that really means either the company, investors, or customers need to make a moral decision at the cost of economic gain or convenience/novelty.
- Silphendio 2y agoChild porn is already heavily "regulated". Producing the stuff is illegal and AI service providers must take steps to prevent it.
- _heimdall 2y agoThis is a blind spot for me, do existing legal protections generally cover fake/generated content? To be clear, the context of this thread is extremely important. I don't know how far our existing laws might go to make sure ML generated child porn is illegal and I'd personally feel much more comfortable to know that ML generation isn't some kind of legal loophole.
- tbrownaw 2y agoLast I heard it varies by jurisdiction. Some only prohibit things that actually involve actual children, some ban based on subject matter even if it's say a cartoon drawing. I assume that training separate models for separate jurisdictions would be rather expensive, so probably unlikely.
- tbrownaw 2y agoNegative reputational or legal consequences directed at the company providing the AI.
- wokwokwok 2y agoI feel like, when companies start venturing into adult entertainment, something has gone wrong. There’s nothing specifically wrong with it, but there are “legit” companies and “kinda shady” ones, and when you start hanging out with a bad crowd… Sad to see. I feel like it really completes the fall from grace for OpenAI.