[upwardbound]

Yeah. I think the most realistically-achievable positive outcome is that Microsoft is forced to give up their new board-observer seat, which seems highly improper for them to have since the board they are observing is supposed to make decisions that benefit all of humanity equally. If Microsoft gets to have a fly on the wall in those discussions, it gives them a gold mine of juicy insider knowledge about which sectors of the economy are about to be affected next by Generative AI — knowledge which they can use to front-run the market by steering the roadmaps of Bing, MS Office, etc. so as to benefit from upcoming OpenAI product launches before any non-insiders are aware of what OpenAI is currently planning.

Microsoft plainly shouldn't be allowed to have this advantage, in that giving an advantage to any one party directly harms the mandate set forth in OpenAI's Charter:

    Broadly distributed benefits

    We commit to use any influence we obtain over AGI’s deployment to ensure it is used for the benefit of all, and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power.

    Our primary fiduciary duty is to humanity. We anticipate needing to marshal substantial resources to fulfill our mission, but will always diligently act to minimize conflicts of interest among our employees and stakeholders that could compromise broad benefit.

https://openai.com/charter

It certainly seems that Microsoft, a "stakeholder", has managed to get a highly improper listening seat that will give them the ability to act on insider information about what's coming next in AI, allowing Microsoft to front-run the rest of the AI software industry and all those industries it affects, in a way that will plainly "compromise broad benefit". (Since any wealth that accrues excessively to Microsoft shareholders is not distributed to other humans who don't hold Microsoft shares.)

A mere 10 days ago, Nadella was shamelessly throwing his weight around on national TV, by appearing on CNBC where he improperly pressured the OpenAI non-profit board — which owes nothing to him legally or morally — to give him more deference, in direct violation of the "always diligently act to minimize conflicts of interest among our employees and stakeholders that could compromise broad benefit" provision of the OpenAI non-profit charter.

https://www.cnbc.com/2023/11/20/microsoft-ceo-nadella-says-o...

[sanderjd]

I doubt Microsoft will be punished for their role in this (it's not clear to me that they did anything wrong...), but I'll be surprised if OpenAI is allowed to still exist as a non-profit a couple years from now.

[_ct0i]

Of course they'll still be allowed to still be a non-profit. If they weren't "allowed" to, that would be a total victory for the Altman-aligned faction that sought to corrupt OpenAI from within and cause it to abandon its charter, and a total loss for Elon Musk and others that donated a total of $133M to the non-profit due to its charter. https://techcrunch.com/2023/05/17/elon-musk-used-to-say-he-p...

Musk and other donors are the most obvious aggrieved parties, and it is also arguably the case that every member of humanity has standing to sue for violation of the charter, because the charter explicitly declares that the primary fiduciary duty of OpenAI, Inc. (which is a non-profit) is to humanity broadly. (Therefore, every human is financially harmed by any charter violation, with such harm manifesting as a reduction in the net present value of the future benefits each human will receive from safe, broadly beneficial AGI.)

The SEC's role in fiduciary misconduct cases is not to rewrite a company's charter - that would be extremely improper and the opposite of their mandate. The SEC's mandate is to be the protectors of the status quo and of the original intent of the organizers of an entity. In this case, that means they will seek to protect the OpenAI non-profit's charter from efforts by Sam and others to erode the charter's power in violation of Sam's fiduciary duty to said charter.

The SEC's job in fiduciary misconduct situations is to remedy the situation by reversing improper governance decisions and forcing the fiduciary (Sam) to uphold their legal duty, which in Sam's case is his contractually bound duty to uphold the Charter of the non-profit entity named OpenAI, Inc. https://en.wikipedia.org/wiki/OpenAI#:~:text=the%20non%2Dpro....

OpenAI states this very clearly in multiple places on their website. For example:

"each director must perform their fiduciary duties in furtherance of its mission—safe AGI that is broadly beneficial" - https://openai.com/our-structure

If you would like to read about the types of remedies available in fiduciary duty violation cases, I recommend this resource:

Book Chapter:

    REMEDIES FOR BREACH OF FIDUCIARY DUTY CLAIMS

https://m.winstead.com/portalresource/lookup/poid/Z1tOl9NPlu...

For example (quoting from the book chapter above):

    C. Permanent Injunction
    A breach-of-fiduciary-duty plaintiff may be entitled to an award of a permanent injunction as a remedy. ...
    The purpose of an injunction is to remove the advantage created by the wrongful act.

In the context of this suit, the permanent injunction or injunctions could block all of the following:

    • Permanently prohibit Microsoft from holding any board seat or board observer seat on the OpenAI board
    • Permanently prohibit Microsoft from making mass employment offers to OpenAI staff (a practice which, in the case of bad faith situations such as this one, is known as "workforce raiding")
    • Permanently prohibit Sam Altman from owning Microsoft stock or derivatives thereof
    • Permanently prohibit Microsoft from receiving early indications of OpenAI research & product roadmaps earlier than the general public
    • Permanently prohibit Sam Altman from receiving job offers at Microsoft or any of its subsidiaries until after a sufficient cooling-off period has elapsed since departing OpenAI, to limit future occurrences of the "revolving door" mechanic documented here:

https://news.ycombinator.com/item?id=38387518

[sanderjd]

I dunno, seems more likely that they'd shutter a nonprofit that's a clear sham and assess a bunch of taxes and penalties on the for-profit subsidiary that was the beneficiary of the sham. But beats me!

[jacquesm]

It was supposed to have two functions: tax dodge + pull the wool over the eyes of regulators. It may fail on both counts now that they've shown that the nonprofit wasn't functional in its stated capacity. Win the battle, possibly lose the war?

[sanderjd]

Yeah this is what I think as well. Certainly regulation won't be avoided now, and regulators will (or should) know not to trust Altman to contribute to developing it.