Hacker News new | ask | show | jobs
by solumunus 221 days ago
I honestly don’t remember many if any people saying that at all, those would have been extremely fringe positions.
3 comments
jvelo 221 days ago
Robert Scoble was saying something of that effect about the upcoming ChatGPT 4 if I remember correctly
link
delusional 221 days ago
From Sam's own blog: "We are now confident we know how to build AGI as we have traditionally understood it."

Another quote: "Trying GPT-4.5 has been much more of a 'feel the AGI' moment among high-taste testers than I expected!"

link
solumunus 221 days ago
So not quite. We’re all aware that some players engage in Musk style marketing.
link
HarHarVeryFunny 221 days ago
It started with GPT-2 - not sure if OpenAI really believed it, or were just hyping it up, but they initially withheld public release of GPT-2 because it was "too powerful and dangerous"...
link
solumunus 221 days ago
Obvious marketing if you ask me.
link
HarHarVeryFunny 221 days ago
Yeah.

At the time there was no obvious reason not to trust that OpenAI was trying to act for the benefit of society, per their charter, so it seemed like an abundance of caution, and this level of LLM capability was new to most of us, so it was hard to guess how dangerous it actually was...

However, in retrospect, seeing how OpenAI continues to behave, it may well have just been to get publicity.

This whole "Be warned, we're about to release something that will destroy society!" shtick seems to be a recurring thing with the AI CEOs, specifically Altman and Amodei (who switched into hardcore salesman mode about a year ago).

The latest Twitter "warning" from Altman is to claim that their AI will soon be at the level of their AI developers, and so we should be prepared (for the self-accelerating singularity I suppose). Maybe this inspired someone to write him another trillion dollar check?

link