owl flight on Nostr: I took a look at OpenAI's website to see what they have to say about "AI safety". The ...
I took a look at OpenAI's website to see what they have to say about "AI safety". The writing is vague and repetitive, and the concerns expressed are generally pretty amorphous. It rather feels like the AI safety talk is being used as a marketing ploy to exeggerate the significance of the product. Elevator safety seems like a more pressing problem.
quoting nevent1q…wwxxI couple of months ago I attended a conference where they talked a lot about "AI safety".
Nobofy bothered to explain what exactly is supposed to be *dangerous* about AI.
When you ask them, they tell you that they are not worried about a "Terminator" scenario (that would be silly).
But if it is not that, then what is it? Does someone know?