OpenAI: AI will kill us. Anthropic: AI will kill us. InflectionAI: AI will kill us. Nvidia:

Video from @chrisprucha >

So once you start making money with AI ,the AI risks disappear. Got it. 👨‍💻👍

Of course! Nvidia is the only one out of that list who benefits from the proliferation of new AI companies

Saying "AGI is science fiction" is a stupid argument, because sci-fi has a habit of becoming reality. We've got AIs winning photography contests, writing code, doing poetry. Pretty sure that was sci-fi just two years ago. Saying "Humans will always decide" is also a stupid argument. What if a terrorist uses AI to design a powerful new pathogen or cyberweapon? And besides, we're giving more control to AIs by the day, even giving autonomous AI agents access to the internet and all sorts of APIs. I think I know what's going on here. It's about money. NVIDIA has the most to lose from AI safety regulation. Stopping the proliferation of the tooling needed for large scale training runs (GPUs) will harm NVIDIA's business. If we choose safety, NVIDIA's stock is harmed.

We get it, they just downplay how powerful AI will end up and pretend we won’t get much more progress than a GPT-6 chatbot. e/acc seems to be doing the same at times. Not sure if it’s lack of imagination or some strategic deceptive play.

It's almost like every book and movie we've seen about AI has used it as a cheap out-group villain for humanity to rally against to excuse poor writing. Really enjoyed Artifice Girl, which is one of the few movies to correct this negative stereotype in media.

Killer #AI is great marketing. Start up companies need this. Established ones don't.

Nvidia has chosen the gun lobby position. There are no bad AI, only bad people with AI. 🤖

I honestly think this is the most realistic perspective. I think humans are incredibly biased towards risk, doom and negative outcomes especially when it comes to unknowns or novel events.