Business ethics professor interested in AI safety (website)
Ewan Kingston
Karma: 10
Ah, that might make sense, from before there was much public information about the Trust? I couldn’t find any evidence of this online but it seems plausible.
Revisiting this excellent thread, with the looming prospect of an Anthropic IPO.
This is my current understanding of the state of play regarding the actual influence of the Long-Term Benefit Trust (hereafter “the Trust”)
1) Does the Trust actually elect a majority of board members, as promised by September 2027?
Kinda.
On April 14 2026, Anthropic announced that Vas Narasimhan was appointed to the board of directors by the trust and “With Narasimhan’s appointment, Trust-appointed directors now make up a majority of the Board.” edit: [announcement]
[This retroactively suggested that Chris Lidell’s Feb 13 2026 appointment was by the Trust, even though this was not detailed in theannouncement, unlikely prior appointments announcements like that of Jay Kreps (announcementpara 3) or Reed Hastings (announcementpara 1)]Edit: it’s more likely that Daniela is being counted as trust-appointed—thanks @Zach Stein-PerlmanHowever, on May 22 Jay Kreps stepped down from the board of directors (LinkedIn announcement)
That leaves six members on the board, three (Narasimhan, Hastings, 1 other) elected by the Trust and three (Dario Amodei, Daniela Amodei and Yasmin Razavi) elected by stockholders. I haven’t seen any timeline for replacing Kreps on the board.
Aside: it still seems strange that the ability of the LTBT to elect a majority of the board is still described in the future tense by Anthropic in their description of the trust that they continue to link to.
2) What perspective does the Trust actually reflect?
In one of his comments [not sure how to link it], OP Zach Stein-Perlman said:
It is maybe quite concerning if they—the two AI safety experts—are gone and not replaced by AI safety experts, and the Trust is putting non-AI-safety people on the board. Also I’m disappointed that Anthropic didn’t cause me to know this before.]
The Trust is now just Neil Buddy Shah, Richard Fontaine, and Ben Bernanke. (Perhaps the size of the trust has been changed by an investor supermajority?) It seems some aspects of AI safety, particularly foreign policy and economic disruption might be well represented, but loss of control/alignment seems to be less so (although Bernanke endorsed If Everyone Builds It).
3) I have a lot more questions, especially:
How much detail of this current structure is likely to be legally required to be revealed for an IPO to proceed (I’m guessing a lot)
How much is likely to change before an IPO (again, probably a lot).
4) The other purpose of posting here is just to check in with other people focused on corporate governance of AI frontier labs to see what you’ve been learning/thinking/wondering.
I’m very new to alignment research: I’m a college professor with a philosophy background trying to write a realistic near-future case study for an undergraduate business ethics class about A(G)I, alignment and safety. I think I want to focus the case study around a (mid 2027?) decision in a particular company to implement or not implement chain-of-continuous-thought reasoning in developing a new, powerful LLM model.
My main questions for this community are:
(1) Am I correct in thinking chain-of-continuous-thought not yet been widely implemented in major LLMs? (i.e), Are the big players still using language tokens for chain-of-thought reasoning rather than vectors?
(2) Is it accurate that the choice to use chain-of-continuous-thought is quite likely to involve a sacrifice of interpretability in return for more (efficient) ability to “remember context” for the LLM?
Thank you on behalf of myself and my students!
That definitely seems most likely—edited the original comment.