Raw LLM Responses

I play a lot with generative AI as a hobby. Nothing like they talk about in the video though. But I know how easy and computationally cheap it is to run a model that will paste a face into a picture or even generate new pictures with a given face. Video would be a little more expensive (again in terms of computation) but still as easy. What should be said is, we are not feeling the whole effect of this technology yet. The fakes that can be created may and will be used to influence people. I was quite surprised that this has not been used in most recent US elections. I think that the stopgap measures would be to get some sort of watermarking into publicly available gen AI or train detection models to discern what is real and what's deepfake. But there will always be a process to circumvent these measures or else the generative model will be unusable for it's purposes. More permanent solution would be to change the public opinion so that any sexual content about a person should not stigmatize or be offensive to that person in any way. The whole situation described in the video is only because it is a great shame having your pics/videos circulating online. But this fruit is only sweet as long as it's forbidden. It should be like: there's a sexual content of you online? So what? Fake or real, it doesn't matter. There's sexual content of everyone and everything, even of that perp who made your nudies having intercourse with a pig. It's only if the whole society gets inoculated against it we will have peace for individuals.