4 ms·

Bulletproofing how?

jacquesm 3y ago
By making sure they had buy-in from all of the major stakeholders to ensure that it wouldn't cause blowback. I think they failed miserably at predicting the kind of shitstorm that this would cause and if they did predict it then they're even more stupid than they already look. Of course this was going to blow up. Note that two founders and shareholders in the non-profit were presumably there when this vote was taken and the fact that those two didn't see eye to eye with the board was waltzed over because it was just a headcount. That's naive, to put it mildly. A move like that takes weeks if not months to prepare and even then you have to be super careful. You don't do that by taking a snap poll and hoping nobody else will notice.
- himaraya 3y ago
  The nonprofit the board oversees doesn't have any shareholders and only answers to its charter. Sam & Greg allegedly sidelined Ilya shortly before all this went down, so Ilya probably felt he had to take drastic action regardless of the blowback.
  - sigil 3y ago
    You seem to be confusing “stakeholders” with “shareholders.”
    - himaraya 3y ago
      Yes, I'll correct that.
      - jacquesm 3y ago
        Well, you can correct it but the 'for profit' arm does have shareholders and those are quite possibly now well positioned to take the board to task for straying outside their mandate by not exercising their duty of care. They will have to have a very good defense from such a challenge.
        himaraya 3y ago
        The for-profit shareholders can only sway the board through the employees who count as stakeholders in the nonprofit, hence the PR campaign. The board's mandate remains upholding the nonprofit's charter—no fiduciary duty to investors to speak of.
        jacquesm 3y ago
        On paper, yes. But good luck with that in the real world, as you can see from the unfolding mess.
        sangnoir 3y ago
        > But good luck with that in the real world, as you can see from the unfolding mess. Any board member who defends a non-profits mission at the cost of being systemically assailed by "Reputation Management" consultants, the venture class and fellow travelers has my respect. I expect to see more hit pieces by pet journalists, blog posts and vituperative opinion-column articles sent in on Sand Hill Road letterheads.
        jacquesm 3y ago
        Maybe. That all depends on whether that mission was really what was at stake or whether they allowed themselves to be roped into some kind of putsch. I've yet to see anything that proves without a doubt that the best remedy for whatever ailed OpenAI was to terminate Sam Altman and to risk the company imploding. It seems a bit drastic, to put it mildly. If it turns out that they were the last ones holding the fort against the hordes I'll be happy to change my tune on that.
- swatcoder 3y ago
  Alternately, it was seen as necessary to fire Altman and blowback was seen as inevitable. Do you think Microsoft, who had close personal connections to Altman and who was a direct and outsized beneficiary of his leadership was just going to be like “Ah, yeah, you guys got a point there. Let’s ditch him.”? Of course not. They would have used their leverage to prevent his ouster, and the board and its charter would be shown as completely impotent. The cause they were formed over would have been lost. The firing may also have ruined the effort, but did show in a public show of principle rather than in being quietly disempowered. You can see where an idealist who thinks there are real stakes to the cause would choose the public firing of Altman over the mufflwd suffocation of their principles.
  - jacquesm 3y ago
    > Alternately, it was seen as necessary to fire Altman and blowback was seen as inevitable. That may well be the case, but I have - yet - to see any evidence of that and as it stands all that I think will happen at this point is that the board will find out where real power is held. Hint: it isn't with them. They're there not because the are free and independent thinkers who will keep AI safe for humanity (even though they may genuinely think so) they are there to serve PR function (to present an image of a set of safe hands to the world) and to serve as a backstop for any kind of regulator that might want to get themselves into the mix. Essentially they're a fig leaf that has now decided that they hold actual power and I think that was a mistake on their part. Unless they can show how whatever Altman was doing directly caused OpenAI's mission to be endangered (and the various stakeholders seem to be extremely conflicted on this) I don't think their decision will stand, and the amount of blowback incurred alone is a good indication of that. In fact I think it will substantially endanger the governance of OpenAI's activities in the longer term. Note that they should have resigned or acted much, much earlier if they felt that OpenAI was not acting true to its mission because I really don't see the difference between all of the stuff that has happened so far compared to what they now claim is the reason for the ouster, it may well be the straw that broke the camel's back but then it wouldn't come out in the way that it did. That mostly seems post-fact reason to try to explain away their blunders. > Do you think Microsoft, who had close personal connections to Altman and who was a direct and outsized beneficiary of his leadership was just going to be like “Ah, yeah, you guys got a point there. Let’s ditch him.”? No, and that's precisely why they made a mistake because it may not be a viable action without that kind of support. A board has a number of policy instruments at its disposal: advisories, votes and resignation. They seem to have believed that after identifying Altman as the main cause of concern a vote on his continued tenure was the appropriate instrument. I think that in a 'simple' non profit, one not encumbered by a for-profit part that also happens to have outsider shareholders (besides employees and execs) that is difficult enough, you need to really have a solid case. But absent a solid case a shotgun vote is about as blunt an instrument as it gets and the use of that instrument is - normally - reserved for crisis situations, such as impending skeletons about to fall out of closets. Then you explain afterwards what was up and everybody will fall in line. That is definitely not happening. And you don't use that approach for someone who is nominally been phenomenally successful at achieving what they set out to do - and that's me speaking as not exactly a friend of Altman's, I think he has ethics issues and that is precisely why in my opinion this was a huge mistake: Sam will likely come out of this as much stronger entrenched than he was before. > Of course not. They would have used their leverage to prevent his ouster, and the board and its charter would be shown as completely impotent. The cause they were formed over would have been lost. Not quite, they could have simply made a statement that they felt that their role as oversight board was made impossible because of the way the governance structure was set up, that they are specifically aware of a number of things that OpenAI is doing that are unacceptable and then to resign. Plenty of precedent for that. > The firing may also have ruined the effort, but did show in a public show of principle rather than in being quietly disempowered. That's true, in the end the result may well be identical. But then they've paved the way for an OpenAI without any oversight at all and this is a major problem. I'd much rather have seen that they had used their influence in a more constructive manner because that would have had a larger chance at making it stick. If Altman prevails here - which has a pretty good chance - OpenAI will be effectively without any controls at all. Because obviously the risk of a bunch of placeholders for stakeholders going rogue on Friday afternoon at half an hour to the bell isn't going to be acceptable to shareholders. > You can see where an idealist who thinks there are real stakes to the cause would choose the public firing of Altman over the mufflwd suffocation of their principles. Yes, I can see that. Doesn't make it the smartest move though.

Bulletproofing how?

By making sure they had buy-in from all of the major stakeholders to ensure that it wouldn't cause blowback. I think they failed miserably at predicting the kind of shitstorm that this would cause and if they did predict it then they're even more stupid than they already look. Of course this was going to blow up. Note that two founders and shareholders in the non-profit were presumably there when this vote was taken and the fact that those two didn't see eye to eye with the board was waltzed over because it was just a headcount. That's naive, to put it mildly. A move like that takes weeks if not months to prepare and even then you have to be super careful. You don't do that by taking a snap poll and hoping nobody else will notice.

himaraya 3y ago
The nonprofit the board oversees doesn't have any shareholders and only answers to its charter. Sam & Greg allegedly sidelined Ilya shortly before all this went down, so Ilya probably felt he had to take drastic action regardless of the blowback.
- sigil 3y ago
  You seem to be confusing “stakeholders” with “shareholders.”
  - himaraya 3y ago
    Yes, I'll correct that.
    - jacquesm 3y ago
      Well, you can correct it but the 'for profit' arm does have shareholders and those are quite possibly now well positioned to take the board to task for straying outside their mandate by not exercising their duty of care. They will have to have a very good defense from such a challenge.
      - himaraya 3y ago
        The for-profit shareholders can only sway the board through the employees who count as stakeholders in the nonprofit, hence the PR campaign. The board's mandate remains upholding the nonprofit's charter—no fiduciary duty to investors to speak of.
        jacquesm 3y ago
        On paper, yes. But good luck with that in the real world, as you can see from the unfolding mess.
        sangnoir 3y ago
        > But good luck with that in the real world, as you can see from the unfolding mess. Any board member who defends a non-profits mission at the cost of being systemically assailed by "Reputation Management" consultants, the venture class and fellow travelers has my respect. I expect to see more hit pieces by pet journalists, blog posts and vituperative opinion-column articles sent in on Sand Hill Road letterheads.
        jacquesm 3y ago
        Maybe. That all depends on whether that mission was really what was at stake or whether they allowed themselves to be roped into some kind of putsch. I've yet to see anything that proves without a doubt that the best remedy for whatever ailed OpenAI was to terminate Sam Altman and to risk the company imploding. It seems a bit drastic, to put it mildly. If it turns out that they were the last ones holding the fort against the hordes I'll be happy to change my tune on that.
swatcoder 3y ago
Alternately, it was seen as necessary to fire Altman and blowback was seen as inevitable. Do you think Microsoft, who had close personal connections to Altman and who was a direct and outsized beneficiary of his leadership was just going to be like “Ah, yeah, you guys got a point there. Let’s ditch him.”? Of course not. They would have used their leverage to prevent his ouster, and the board and its charter would be shown as completely impotent. The cause they were formed over would have been lost. The firing may also have ruined the effort, but did show in a public show of principle rather than in being quietly disempowered. You can see where an idealist who thinks there are real stakes to the cause would choose the public firing of Altman over the mufflwd suffocation of their principles.
- jacquesm 3y ago
  > Alternately, it was seen as necessary to fire Altman and blowback was seen as inevitable. That may well be the case, but I have - yet - to see any evidence of that and as it stands all that I think will happen at this point is that the board will find out where real power is held. Hint: it isn't with them. They're there not because the are free and independent thinkers who will keep AI safe for humanity (even though they may genuinely think so) they are there to serve PR function (to present an image of a set of safe hands to the world) and to serve as a backstop for any kind of regulator that might want to get themselves into the mix. Essentially they're a fig leaf that has now decided that they hold actual power and I think that was a mistake on their part. Unless they can show how whatever Altman was doing directly caused OpenAI's mission to be endangered (and the various stakeholders seem to be extremely conflicted on this) I don't think their decision will stand, and the amount of blowback incurred alone is a good indication of that. In fact I think it will substantially endanger the governance of OpenAI's activities in the longer term. Note that they should have resigned or acted much, much earlier if they felt that OpenAI was not acting true to its mission because I really don't see the difference between all of the stuff that has happened so far compared to what they now claim is the reason for the ouster, it may well be the straw that broke the camel's back but then it wouldn't come out in the way that it did. That mostly seems post-fact reason to try to explain away their blunders. > Do you think Microsoft, who had close personal connections to Altman and who was a direct and outsized beneficiary of his leadership was just going to be like “Ah, yeah, you guys got a point there. Let’s ditch him.”? No, and that's precisely why they made a mistake because it may not be a viable action without that kind of support. A board has a number of policy instruments at its disposal: advisories, votes and resignation. They seem to have believed that after identifying Altman as the main cause of concern a vote on his continued tenure was the appropriate instrument. I think that in a 'simple' non profit, one not encumbered by a for-profit part that also happens to have outsider shareholders (besides employees and execs) that is difficult enough, you need to really have a solid case. But absent a solid case a shotgun vote is about as blunt an instrument as it gets and the use of that instrument is - normally - reserved for crisis situations, such as impending skeletons about to fall out of closets. Then you explain afterwards what was up and everybody will fall in line. That is definitely not happening. And you don't use that approach for someone who is nominally been phenomenally successful at achieving what they set out to do - and that's me speaking as not exactly a friend of Altman's, I think he has ethics issues and that is precisely why in my opinion this was a huge mistake: Sam will likely come out of this as much stronger entrenched than he was before. > Of course not. They would have used their leverage to prevent his ouster, and the board and its charter would be shown as completely impotent. The cause they were formed over would have been lost. Not quite, they could have simply made a statement that they felt that their role as oversight board was made impossible because of the way the governance structure was set up, that they are specifically aware of a number of things that OpenAI is doing that are unacceptable and then to resign. Plenty of precedent for that. > The firing may also have ruined the effort, but did show in a public show of principle rather than in being quietly disempowered. That's true, in the end the result may well be identical. But then they've paved the way for an OpenAI without any oversight at all and this is a major problem. I'd much rather have seen that they had used their influence in a more constructive manner because that would have had a larger chance at making it stick. If Altman prevails here - which has a pretty good chance - OpenAI will be effectively without any controls at all. Because obviously the risk of a bunch of placeholders for stakeholders going rogue on Friday afternoon at half an hour to the bell isn't going to be acceptable to shareholders. > You can see where an idealist who thinks there are real stakes to the cause would choose the public firing of Altman over the mufflwd suffocation of their principles. Yes, I can see that. Doesn't make it the smartest move though.