9 ms·

It's just so strange. This is such a clearly justifiable reason that the fact that they didn't argue it... or argue.. anything, makes me very suspicious that it

by drooby 3y ago

It's just so strange. This is such a clearly justifiable reason that the fact that they didn't argue it... or argue.. anything, makes me very suspicious that it is correct.

7e 3y ago
They were either scared of being sued for defamation or unwilling to divulge existential company secrets. Or both.
- fastball 3y ago
  Doesn't really make sense to be unwilling to divulge company secrets if you're willing to gut the company for this hill.
  - 0xDEAFBEAD 3y ago
    They weren't willing to gut the company. That's why Sam is back as CEO.
    - jxi 3y ago
      It sounded like they would if they could (for instance trying to sell to Anthropic or instating a "slow it way down" CEO), but they even failed at that. Not an effective board at all.
      - 0xDEAFBEAD 3y ago
        >for instance trying to sell to Anthropic or instating a "slow it way down" CEO I wouldn't put these in the same category as "90% of staff leaves for Microsoft". In any case, let's not confuse unsuccessful with incompetent. (Or incompetent with immoral, for that matter.)
    - jacquesm 3y ago
      They were willing but failed, and mostly on account of not doing enough prepwork.
  - 6gvONxR4sf7o 3y ago
    It's remarkable that the old board is the side characterized as willing to blow up the company, since it was Altman's side who threatened to blow it up. All the old board really did was fire Altman and remove Brockman from the board.
  - DebtDeflation 3y ago
    Exactly. It sounds like those board members themselves were acting in the interest of profit instead of the "benefit all humanity" mission stuff, no different than Altman. If anything then, the only difference between the two groups is one of time horizon. Altman wants to make money from the technology now. The board wants to wait until it's advanced enough to take over the world, and then take over the world with it. For the world's benefit, of course.
- 0xDEAFBEAD 3y ago
  I think "unwilling to divulge company secrets" is the best explanation here. We know that OpenAI does a staged release for their models with pre-release red-teaming. Helen says the key issue was the board struggling to "effectively supervise the company": https://nitter.net/hlntnr/status/1730034017435586920#m https://nitter.net/hlntnr/status/1730034017435586920#m Here's Nathan Labenz on how sketchy the red-teaming process for GPT4 was. Nathan states that OpenAI shut him out soon after he reached out to the board to let them know that GPT4 was a big deal and the board should be paying attention: https://nitter.net/labenz/status/1727327424244023482#m https://nitter.net/labenz/status/1727327424244023482#m [Based on the thread it seems like he reached out to people outside OpenAI in a way which could have violated a confidentiality agreement -- that could account for the shutout] My suspicion is that there was a low-level power struggle ongoing on the board for some time, but the straw that broke the camel's back was something like Nathan describes in his thread. To be honest I don't understand why his thread is getting so little play. It seems like a key piece of the puzzle. In any case, I don't think it would've been right for Helen to say publicly that "we hear GPT-5 is lit but Sam isn't letting us play with it", since "GPT-5 is lit" would be considered confidential information that she shouldn't unilaterally reveal.
  - g42gregory 3y ago
    So what is Nathan Labenz saying? That GPT-4 is dangerous somehow? It will get many people out of jobs? MS Office got all the typists out of jobs. OCR and Medical Software got all the medical transcriptionists out of jobs. And they created a lot more jobs in the process. GPT-4 is a very powerful tool. It has not a whiff of AGI in it. The whole AGI "scare" seems to be extremely political.
    - 0xDEAFBEAD 3y ago
      Nathan says the initial version of GPT-4 he red-teamed was "totally amoral" and it was happy to plan assassinations for him: https://nitter.net/labenz/status/1727327464328954121#m https://nitter.net/labenz/status/1727327464328954121#m Reducing the cost of medical transcription to ~$0 is one thing. Reducing the cost of assassination to ~$0 is quite another.
sanderjd 3y ago
Yeah, I totally agree! Like, this is such an obviously true and valid reason to fire him, but they never came out and said it! So ... is this not what it actually was? Or ... what? It truly is mystifying.
- danbmil99 3y ago
  From a doomer/EA perspective, publicly saying that GPT5 is AGI or such would likely inspire & accelerate labs around the world to catch up. Thus it was more "Altruistic" / aligned with humanity's fate to stay mum and leave the situation cloudy.
  - esafak 3y ago
    No, it would not; they are already trying to catch up.
    - brhsagain 3y ago
      Having an existing example showing that something difficult is possible causes everyone else to replicate it much faster, like how a bunch of people started running sub-4-minute miles after the first guy did it.
    - throwuwu 3y ago
      If you know the outcome is favourable then you go all in. Right now the other competitors are just trying to match GPT4, if they knew AGI was achievable then they would throw everything they have at it in order to not be left out.
      - xp84 3y ago
        Show me the researchers and companies where the people in charge and the people doing the work don’t think it’s possible to get better than GPT-4 and are slow-rolling things. I suppose maybe there are ones where they are slow-rolling because of their opinions about existential risk of AGI. But that’s not contingent on what OpenAI says or does.

by drooby 3y ago

It's just so strange. This is such a clearly justifiable reason that the fact that they didn't argue it... or argue.. anything, makes me very suspicious that it is correct.

7e 3y ago

They were either scared of being sued for defamation or unwilling to divulge existential company secrets. Or both.

fastball 3y ago
Doesn't really make sense to be unwilling to divulge company secrets if you're willing to gut the company for this hill.
- 0xDEAFBEAD 3y ago
  They weren't willing to gut the company. That's why Sam is back as CEO.
  - jxi 3y ago
    It sounded like they would if they could (for instance trying to sell to Anthropic or instating a "slow it way down" CEO), but they even failed at that. Not an effective board at all.
    - 0xDEAFBEAD 3y ago
      >for instance trying to sell to Anthropic or instating a "slow it way down" CEO I wouldn't put these in the same category as "90% of staff leaves for Microsoft". In any case, let's not confuse unsuccessful with incompetent. (Or incompetent with immoral, for that matter.)
  - jacquesm 3y ago
    They were willing but failed, and mostly on account of not doing enough prepwork.
- 6gvONxR4sf7o 3y ago
  It's remarkable that the old board is the side characterized as willing to blow up the company, since it was Altman's side who threatened to blow it up. All the old board really did was fire Altman and remove Brockman from the board.
- DebtDeflation 3y ago
  Exactly. It sounds like those board members themselves were acting in the interest of profit instead of the "benefit all humanity" mission stuff, no different than Altman. If anything then, the only difference between the two groups is one of time horizon. Altman wants to make money from the technology now. The board wants to wait until it's advanced enough to take over the world, and then take over the world with it. For the world's benefit, of course.
0xDEAFBEAD 3y ago
I think "unwilling to divulge company secrets" is the best explanation here. We know that OpenAI does a staged release for their models with pre-release red-teaming. Helen says the key issue was the board struggling to "effectively supervise the company": https://nitter.net/hlntnr/status/1730034017435586920#m https://nitter.net/hlntnr/status/1730034017435586920#m Here's Nathan Labenz on how sketchy the red-teaming process for GPT4 was. Nathan states that OpenAI shut him out soon after he reached out to the board to let them know that GPT4 was a big deal and the board should be paying attention: https://nitter.net/labenz/status/1727327424244023482#m https://nitter.net/labenz/status/1727327424244023482#m [Based on the thread it seems like he reached out to people outside OpenAI in a way which could have violated a confidentiality agreement -- that could account for the shutout] My suspicion is that there was a low-level power struggle ongoing on the board for some time, but the straw that broke the camel's back was something like Nathan describes in his thread. To be honest I don't understand why his thread is getting so little play. It seems like a key piece of the puzzle. In any case, I don't think it would've been right for Helen to say publicly that "we hear GPT-5 is lit but Sam isn't letting us play with it", since "GPT-5 is lit" would be considered confidential information that she shouldn't unilaterally reveal.
- g42gregory 3y ago
  So what is Nathan Labenz saying? That GPT-4 is dangerous somehow? It will get many people out of jobs? MS Office got all the typists out of jobs. OCR and Medical Software got all the medical transcriptionists out of jobs. And they created a lot more jobs in the process. GPT-4 is a very powerful tool. It has not a whiff of AGI in it. The whole AGI "scare" seems to be extremely political.
  - 0xDEAFBEAD 3y ago
    Nathan says the initial version of GPT-4 he red-teamed was "totally amoral" and it was happy to plan assassinations for him: https://nitter.net/labenz/status/1727327464328954121#m https://nitter.net/labenz/status/1727327464328954121#m Reducing the cost of medical transcription to ~$0 is one thing. Reducing the cost of assassination to ~$0 is quite another.

sanderjd 3y ago

Yeah, I totally agree! Like, this is such an obviously true and valid reason to fire him, but they never came out and said it! So ... is this not what it actually was? Or ... what? It truly is mystifying.

danbmil99 3y ago
From a doomer/EA perspective, publicly saying that GPT5 is AGI or such would likely inspire & accelerate labs around the world to catch up. Thus it was more "Altruistic" / aligned with humanity's fate to stay mum and leave the situation cloudy.
- esafak 3y ago
  No, it would not; they are already trying to catch up.
  - brhsagain 3y ago
    Having an existing example showing that something difficult is possible causes everyone else to replicate it much faster, like how a bunch of people started running sub-4-minute miles after the first guy did it.
  - throwuwu 3y ago
    If you know the outcome is favourable then you go all in. Right now the other competitors are just trying to match GPT4, if they knew AGI was achievable then they would throw everything they have at it in order to not be left out.
    - xp84 3y ago
      Show me the researchers and companies where the people in charge and the people doing the work don’t think it’s possible to get better than GPT-4 and are slow-rolling things. I suppose maybe there are ones where they are slow-rolling because of their opinions about existential risk of AGI. But that’s not contingent on what OpenAI says or does.