News for hackers. Stuff that matters.
Re:Three senior researchers have resigned from OpenAI by thr8976 on Saturday November 18, 2023 @02:16AM (#38316461)
Aleksander in particular is deeply invested in AI safety as a mission. It's a very confusing departure, since most of the reporting so far indicates that Ilya and the board fired Sam to prioritize safety and non-profit objectives. A huge loss for OpenAI nonetheless.
Re:Three senior researchers have resigned from OpenAI by cinntaile on Saturday November 18, 2023 @02:32AM (#38316561)
The way Sam and Greg were fired maybe led him to no longer have faith in the company and so he quit?
Re:Three senior researchers have resigned from OpenAI by mannyv on Saturday November 18, 2023 @03:00AM (#38316749)
More like the guy who engineered this situation is an asshole and they don't want to work for him.
Re:Three senior researchers have resigned from OpenAI by Obscurity4340 on Saturday November 18, 2023 @03:19AM (#38316876)
Who's the situation-engineer for some of us duller but curious folks?
Re:Three senior researchers have resigned from OpenAI by kcb on Saturday November 18, 2023 @06:09AM (#38318089)
It's been confirmed to be Ilya.
Re:Three senior researchers have resigned from OpenAI by notRobot on Saturday November 18, 2023 @03:21AM (#38316892)
Who was that? How are they an asshole?
Re:Three senior researchers have resigned from OpenAI by deneas on Saturday November 18, 2023 @03:30AM (#38316951)
Important detail: Only Sam was fired, Greg was removed from the board and then later quit. Source: https://twitter.com/gdb/status/1725667410387378559
Re:Three senior researchers have resigned from OpenAI by MattRix on Saturday November 18, 2023 @02:37AM (#38316591)
Perhaps you could argue that he wants to stick with Sam and the others because if they start a company that competes with OpenAI, there’s a real chance they catch up and surpass OpenAI. If you really want to be a voice for safety, it’ll be most effective if you’re on the winning team.
Re:Three senior researchers have resigned from OpenAI by Closi on Saturday November 18, 2023 @03:09AM (#38316805)
Depends how much research is driven by Ilya…
Re:Three senior researchers have resigned from OpenAI by I_am_uncreative on Saturday November 18, 2023 @03:14AM (#38316843)
I dunno, the moat Sam tried to build might make it hard to make a competitor.
Re:Three senior researchers have resigned from OpenAI by jatins on Saturday November 18, 2023 @03:33AM (#38316974)
We are about to find out if the moats are indeed that strong.

xAI recently showed that training a decent-ish model is now a multi-month effort. Granted GPT-4 is still farther along than others but curious how many months/resources does that add up when you have the team that built it in the first place

But also, starting another LLM company might be too obvious a thing to do. Maybe Sam has another trick up his sleeve? Though I suspect he is sticking with AI one way or the other

Re:Three senior researchers have resigned from OpenAI by thekoma on Saturday November 18, 2023 @03:50AM (#38317084)
One funny detail is that the OpenAI charter states that, if this happens, they will stop their own work and help the organisation that is closest to achieving OpenAI's stated goal.
Re:Three senior researchers have resigned from OpenAI by aryamaan on Saturday November 18, 2023 @04:42AM (#38317467)
really?
Re:Three senior researchers have resigned from OpenAI by thekoma on Saturday November 18, 2023 @05:36AM (#38317865)
https://openai.com/charter

Second paragraph of the "Long-term safety" section.

Re:Three senior researchers have resigned from OpenAI by trallnag on Saturday November 18, 2023 @04:59AM (#38317567)
Maybe Sam wants to build something for profit?
Re:Three senior researchers have resigned from OpenAI by cma on Saturday November 18, 2023 @08:26AM (#38319305)
But now it may be the regulations they've gotten in place will make it harder for any new upstarts to approach them.
Re:Three senior researchers have resigned from OpenAI by visarga on Saturday November 18, 2023 @05:17AM (#38317710)
> If you really want to be a voice for safety, it’ll be most effective if you’re on the winning team.

If an AI said that, we'd be calling it "capability gain" and think it's a huge risk.

Re:Three senior researchers have resigned from OpenAI by sundarurfriend on Saturday November 18, 2023 @03:22AM (#38316896)
> since most of the reporting so far indicates that Ilya and the board fired Sam to prioritize safety and non-profit objectives

With evidence, or is this the kind of pure speculation that media indulges in when they have no information and have to appear knowledgeable?

Re:Three senior researchers have resigned from OpenAI by beowulfey on Saturday November 18, 2023 @03:28AM (#38316940)
Twitter rumors from “insiders”
Re:Three senior researchers have resigned from OpenAI by kcb on Saturday November 18, 2023 @06:09AM (#38318094)
No. Statements from Ilya himself.
Re:Three senior researchers have resigned from OpenAI by visarga on Saturday November 18, 2023 @05:05AM (#38317615)
> most of the reporting so far indicates that Ilya and the board fired Sam to prioritize safety and non profit objectives

Maybe Ilya discovered something as head of AI safety research, something bad, and they had to act on it. From the outside it looks as if they are desperately trying to gain control. Maybe he got confirmation that LLMs are a little bit conscious, LOL. No, I am not making this up: https://twitter.com/ilyasut/status/1491554478243258368

Re:Three senior researchers have resigned from OpenAI by Davidzheng on Saturday November 18, 2023 @11:40AM (#38321369)
lol sorry if this is clearly a joke but who cares if it's a little bit conscious. So are fucking pigeons.
Re:Three senior researchers have resigned from OpenAI by visarga on Monday November 20, 2023 @01:07AM (#38343165)
it would be funny if Ilya followed the ranks of Blake Lemoine and went off for AI consciousness
Re:Three senior researchers have resigned from OpenAI by ianbicking on Saturday November 18, 2023 @10:43AM (#38320751)
A rudder only works as long as you are moving faster than the current. I can imagine (some) people concerned with safety also feeling a sense of urgency, because their ability to steer the AI toward the good is limited by their organization's engine of progress.