My reaction to this, and all other discussions on the topic, is that it rings hollow.
Here’s why & I’m very curious for your response:
Conflict of interest is a fact of life in elite, cutting edge work.
“No conflict, no interest” is as old as SV and VC for that reason. It’s a small pool of people who create great companies and invest.
How someone structures themselves around that conflict is the signal. Disclosing potential conflicts of interest is the ethical bar and that has been done.
Whatever the real reason for the original OpenAI’s board to ouster Sam, the current board has found no wrongdoing on Sam’s part and crucially made no request for him to cease investment activity.
My hypothesis: that’s because Sam’s investment is a powerful signal and one the AI startup community greatly benefits from.
I do think Sam’s virtue signaling before Congress has painted Sam into a corner here, but we keep shining the flashlight on it and coming out with the same conclusion: Sam is studious about governance and an astute risk taker. Even the latest ScarJo incident shows that. I don’t see how OpenAI could do better for CEO.
It is possible for Sam to take both positions.
1. Not taking financial gains from OpenAI so that his decisions are guided by building safe AGI
2. Investing in companies that are betting on trends he has the most conviction and knowledge about
Perhaps a simple test is- is the moral hazard here worse than owning stock in OpenAI?
If we don’t expect luminaires to be anything other than people, the OpenAI team’s convergence to bring Sam back is the clearest signal anyone could have on who they believe the best leader to bet on is.
It’s human nature to find fault and we all have our faults. My concern with making this the hill to die on in critiquing Sam & OpenAI is we feed a narrative that has little explanatory power and perpetuates unrealistic expectations on how life in a position of power works.
It’s human nature to find fault and we all have our faults. My concern with making this the hill to die on in critiquing Sam & OpenAI is we feed a narrative that has little explanatory power and perpetuates unrealistic expectations on how life in a position of power works.
Isn't then the criticism about the acceptance that "life in a position of power works" this way and is essentially morally corrupt? Couldn't it be better? Greed is among the chief reasons for such conflicts of interest to arise, shouldn't we criticise such aspect and expect better from the "elites" in positions of power? Since greed will eventually corrupt we could probably want to expect better from the ones taking positions of power, I personally wouldn't like to accept that this is the best system we can live under, and for it to change we need to criticise it for the optics to change over time.
It was once accepted to own slaves, eventually we all landed on the same page: it is morally corrupt to do it, it's unethical. More modern examples like insider trading, stock fraud, were once accepted until we didn't accept it anymore as "it's the way it is".
I don't see a reason to not strive making other morally corrupt aspects of our systems (such as conflicts of interest) also unethical and frowned upon. People are people and shared ethics is how we level ourselves against each other.
How someone structures themselves around that conflict is the signal. Disclosing potential conflicts of interest is the ethical bar and that has been done.
This is kind of the problem though, it hasn't been done. Graham, my example in the post, didn't know YC invested in OpenAI. The previous board didn't know about Altman's ownership of the Startup Fund. His growing investments were, for most of them, published on the WSJ — yesterday.
Whatever the real reason for the original OpenAI’s board to ouster Sam, the current board has found no wrongdoing on Sam’s part and crucially made no request for him to cease investment activity.
The original reason was a lack of trust, and the original board was pushed out so he could return. Of course the current board, one built from it's foundation to be much more receptive and docile to Sam's activities and statements.
It is possible for Sam to take both positions.
1. Not taking financial gains from OpenAI so that his decisions are guided by building safe AGI
2. Investing in companies that are betting on trends he has the most conviction and knowledge about
The whole point here is that the second contradicts the first. If you're still directly making money off OpenAI's decisions & deals, the hand-waving of "being guided by building safe AGI" and not money ceases to be trust, because there are clearly billions at stake for Altman.
we feed a narrative that has little explanatory power and perpetuates unrealistic expectations on how life in a position of power works.
The thing here is, Sam could realistically be an altruistic monk, not driven by any financial incentive when it comes to AI. There are a ton of different options, if he wants to live up to the standard he set for himself.
For example, putting investments into a trust that have to do with AI, not investing in related companies at all, putting a majority of his wealth into a Bill & Malinda Gates-style charity — the list goes on.
If had not tried to tell everyone he's doing this for the good of humanity and not money, he could take a multiple million $ salary + a huge personal stake in OpenAI and no one would bat an eye. It's the fact that he himself is position his image to be one thing, which is not the case.
Comments
Thanks for the open invitation!
My reaction to this, and all other discussions on the topic, is that it rings hollow.
Here’s why & I’m very curious for your response:
Conflict of interest is a fact of life in elite, cutting edge work.
“No conflict, no interest” is as old as SV and VC for that reason. It’s a small pool of people who create great companies and invest.
How someone structures themselves around that conflict is the signal. Disclosing potential conflicts of interest is the ethical bar and that has been done.
Whatever the real reason for the original OpenAI’s board to ouster Sam, the current board has found no wrongdoing on Sam’s part and crucially made no request for him to cease investment activity.
My hypothesis: that’s because Sam’s investment is a powerful signal and one the AI startup community greatly benefits from.
I do think Sam’s virtue signaling before Congress has painted Sam into a corner here, but we keep shining the flashlight on it and coming out with the same conclusion: Sam is studious about governance and an astute risk taker. Even the latest ScarJo incident shows that. I don’t see how OpenAI could do better for CEO.
It is possible for Sam to take both positions.
1. Not taking financial gains from OpenAI so that his decisions are guided by building safe AGI
2. Investing in companies that are betting on trends he has the most conviction and knowledge about
Perhaps a simple test is- is the moral hazard here worse than owning stock in OpenAI?
If we don’t expect luminaires to be anything other than people, the OpenAI team’s convergence to bring Sam back is the clearest signal anyone could have on who they believe the best leader to bet on is.
It’s human nature to find fault and we all have our faults. My concern with making this the hill to die on in critiquing Sam & OpenAI is we feed a narrative that has little explanatory power and perpetuates unrealistic expectations on how life in a position of power works.
Isn't then the criticism about the acceptance that "life in a position of power works" this way and is essentially morally corrupt? Couldn't it be better? Greed is among the chief reasons for such conflicts of interest to arise, shouldn't we criticise such aspect and expect better from the "elites" in positions of power? Since greed will eventually corrupt we could probably want to expect better from the ones taking positions of power, I personally wouldn't like to accept that this is the best system we can live under, and for it to change we need to criticise it for the optics to change over time.
It was once accepted to own slaves, eventually we all landed on the same page: it is morally corrupt to do it, it's unethical. More modern examples like insider trading, stock fraud, were once accepted until we didn't accept it anymore as "it's the way it is".
I don't see a reason to not strive making other morally corrupt aspects of our systems (such as conflicts of interest) also unethical and frowned upon. People are people and shared ethics is how we level ourselves against each other.
Thanks for your reply! Couple thoughts:
This is kind of the problem though, it hasn't been done. Graham, my example in the post, didn't know YC invested in OpenAI. The previous board didn't know about Altman's ownership of the Startup Fund. His growing investments were, for most of them, published on the WSJ — yesterday.
The original reason was a lack of trust, and the original board was pushed out so he could return. Of course the current board, one built from it's foundation to be much more receptive and docile to Sam's activities and statements.
The whole point here is that the second contradicts the first. If you're still directly making money off OpenAI's decisions & deals, the hand-waving of "being guided by building safe AGI" and not money ceases to be trust, because there are clearly billions at stake for Altman.
The thing here is, Sam could realistically be an altruistic monk, not driven by any financial incentive when it comes to AI. There are a ton of different options, if he wants to live up to the standard he set for himself.
For example, putting investments into a trust that have to do with AI, not investing in related companies at all, putting a majority of his wealth into a Bill & Malinda Gates-style charity — the list goes on.
If had not tried to tell everyone he's doing this for the good of humanity and not money, he could take a multiple million $ salary + a huge personal stake in OpenAI and no one would bat an eye. It's the fact that he himself is position his image to be one thing, which is not the case.