Sora 2 is also unbelievably expensive, costing OpenAI at least $5 per 10 second video - and likely more considering Sora 2 also generates synced-up audio. OpenAI loses one subscriber's revenue *every four images generated*. It's unbelievably stupid.
www.wheresyoured.at/sora2-openai...

And, to restate, I'm deliberately staying out of what *I personally* think of kink or contentious content, because I want to talk about how it could be practically enforced. (I don't think it can, basically, by sheer volume, but it can be attempted in a way to minimize obvious harm.)

Any code of conduct that has to be given to a person to moderate has to be understood for it to be useful. Considering human-moderation is notoriously psychologically damaging, some kind of automated moderation could be useful, but *also* loses that nuance of meaning.

I don't think bsky has the perfect answer. I don't think *anyone* has the perfect answer. I do think this is a clearer delineation between cases a moderator can say would be acceptable, and cases where it isn't, and that is *what people had asked for before*.

And while kink *can* indeed be great, therapeutic, what is healing to one can be triggering, or even encouraging, to another. Yes, self-moderation can be enforced, but that's likely to lead to a lot of echo chambers and segmenting. It's not just the target audience, it's periphery.

One of the reasons IG, chatgpt and other AI have been on *blast* lately is their perpetuating of self-harm, ED, and ideation content. Giving bsky the grounds to say 'this is encouraging it' and 'this is supporting understanding and recovery' sounds closer to clear delineation than censorship.

So... please don't be mad or angry. I want to discuss this, and I'm not leading whether I agree or don't agree with you. I want to understand.

The way I'm reading this, bsky is making carveouts for content that is a middle-of-the-road positive way of discussing triggering content.