In October 2023, I learned that Sam Altman’s sister, Annie, claimed that Sam sexually abused her. I wrote a LessWrong post detailing her claims.
My X (formerly Twitter) account: https://x.com/pythagoras5015
Old usernames that I had on LessWrong: pl5015, prometheus5015
Old username that I had on Twitter: prometheus5015 (see my reasons for changing it here: https://x.com/pythagoras5015/status/1860644689109569903)
My mistake. Fixed it. Thanks for pointing that out!
I think you make multiple valid points which are similar to the points I’ve made in my post, but I do think our stances differ in a few ways.
I think that you are certainly correct that psychosis, or a similar type of mental illness / disorder, is a plausible explanatory hypothesis for Annie making the claims that she has.
However, though I do recognize that the simplicity of a hypothesis is a boon to its plausibility, I do not share your belief that we have been unknowingly subsumed by the “MeToo world order”, which has damaged our rationalism and obstructed our ability to recognize this as being obviously the simplest hypothesis. (Though perhaps this is a overly dramatic / inaccurate representation of your assertion.)
While I do agree that this post may encapsulate behavior representative of a person suffering from psychosis, or a similar mental illness, I see the hypothesis space as primarily dual, where mental illness / misrepresentation-of-reality-type hypotheses form one primary subspace, but there exists another primary subspace wherein the behavior detailed in this post is indeed representative of a person who has gone through the things which Annie has claimed she has.
I do appreciate your inclusion of quantitative rates; I think your analysis benefits from it.
Yo—sorry, I meant to keep this in my drafts. I was in the process of making a bunch of edits to my original post, and used this as a sort of “snapshot in time” of a certain segment of my original post to help clear out the the LW editor for my original post while I was modifying it.
Sorry your comment got wasted here.
I’ll be replying to your comment on my main post shortly (which I think is pretty rational, though I do think you may be extrapolating a bit on the MeToo-type influence.)
Thanks!
Actually, right now, I believe that, based upon the information I currently have, it is improper for me to conclude that Sam Altman abused Annie Altman, and that the proper stance is I do not know if Annie Altman’s claims are correct or not; therefore, it is only rational to hold Sam Altman innocent.
However—I’m in the process of gathering more information. Once I’ve conducted research to a degree I consider satisfactory, I’d be happy to hear your reasoning if, at that point, our conclusions disagree. For now, I’ll suggest that you wait until I finish up my research, though feel free to ignore this suggestion if you want :)
The points you make are valid. You also make a good point about the importance of additional context.
I think I may have miscommunicated myself to some extent, based on the fact that I largely agree with your reply here.
The most clear, and most general framing of my motives is this:
My overarching, most fundamental desire is for humanity to have a positive AI future.
Because of this, I want to do my best to determine the validity of a claim(s) such as Annie’s that asserts that the CEO of the world’s (leading) artificial intelligence company / research org / lab / whatever you want to call it may actually be a person of highly questionable morals. The whole reason we got OpenAI in the first place is, apparently, because Elon freaked out when Larry Page called him a ‘specist’ back in 2013. (I will not bother commenting on whether or not I think this was ultimately a good thing. ) I very much want the person leading the development of and (attempts at) alignment of superintelligence to be a good person.
The reason I have made this post here is because of (2), not because I thought that this forum was the right place to worry about the mental health of Annie Altman. While obviously I am concerned for Annie Altman herself independent of my superintelligence / Sam Altman / OpenAI concerns, the reason why I am posting “about Annie” here on LessWrong is because of the potential ramifications of what she is saying about Sam Altman. This isn’t an “Annie Altman post”; it’s a “Sam Altman post” where Annie Altman is the conduit.
Hopefully this framing of mine is more reasonable. And thank you for the compliment—I am trying my best to conduct myself rationally :)
My motivation is pure. I am trying to (rationally) figure out the truth. Though, I’d be epistemologically naive if I expected you to believe me just because I told you “I’m a good person, trust me!”.
Also—I could care less about what people opine (without backing logical/rational arguments.) I could have chosen to do a big long rant with a bunch of clickbait-y quips and half-truthisms on X to try to jack up engagement and suck ad revenue out of X like a leach, but luckily I’m not an asshole (in my humble opinion, lol), so I came here instead. (Not to imply that you said that; I just say this more in an attempt to convey my motives and character.) I came to this site in particular because:
I thought its users would probably understand the significance of a claim that Sam Altman has been quietly hiding the fact that he sexually assaulted his 4-year-old sister.
I thought that its users would be good at calling me out on any logical/irrational bullshit that I (unintentionally) propagated. I want to be right, not to feel right. Say what you will about LessWrong, but its users do love to be quite exacting in their arguments about whether or not they think a person is making rational arguments. Indeed, I’ve modified this post, and my replies, many times in response to comments I’ve received in a way that I think has been to the benefit of the clarity of this post and its conveyance of my position. I’m glad that my karma score has jumped all over the place as I’ve updated my post—it means that LessWrong users are actually thinking critically about the degree to which I am being rational.
It seems to me, at this point, one of two things is true:
Annie Altman is lying left, right,. and center, or is deluded, disconnected from reality, or just misinformed/misunderstanding things to the point that she believes she is telling the truth when she is not.
She is not lying (at least, to some degree.)
Yes, I know we can wonder about base rates and what mental illness we think she may likely have or not have. And such discussions are valid. But I am more interested in (more) concrete research, at the moment, which I’m still working on.
This post is not yet done.
Btw., you don’t have to agree with my (developing) interpretations here. The thing I think is most relevant about this post is the collection of information I’ve assembled, which has nothing to do with my interpretation of it.
Update: While I don’t consider this evidence of a widespread shadowbanning effort, some commenters on Hacker News claim that a post regarding Annie’s claims that Sam sexually assaulted her at age 4 has been being repeatedly removed.
I have updated this post to include this information as well (c.f. item 3.a. in “What Annie has stated on her X account.”)
Good point. I don’t currently know that rate, but agree that it would be helpful in analyzing this matter.
Yes, I think you raise valid points. Given that Annie’s (purported) sexual abuse occurred so long ago, I agree that it is unlikely that, at this point, direct evidence of Sam’s (purported) sexual abuse of her would be able to gathered.
Deviating a bit from your reply to the more general question of “What direct evidence could be provided (e.g. by Annie) to corroborate the claims Annie is making?”—I do think that a potentially useful piece of evidence that could be provided to corroborate (some of) Annie’s claims would be proof that:
Annie’s father left her money in his will.
Annie did not receive this money, as specified in the will.
I understand your concerns, and appreciate your note that you are not accusing me of engaging in improper actions.
Your points are valid. I do acknowledge that the circumstances under which I am making this post, as well as my various departures from objective writing—that is, the instances in this post in which I depart from {solely providing information detailing what Annie has claimed—naturally raise concerns about the motives driving my creation of this post.
I will say:
Regarding the fact that this is my first LessWrong post—I acknowledge that this is unfortunate considering the gravity of the issue which this post addresses.
Regarding my anonymity—I purposefully chose to make this post anonymously. This post discusses a very, very serious topic—the fact that Sam Altman’s sister, Annie Altman, is claiming that he has severely (e.g. sexually) abused her. If Annie’s claims turn out to be (provably) true, this would likely warrant an immediate dismissal of Sam Altman from his current position position as CEO of OpenAI, as well as from a variety of other impactful positions he currently holds. Given the gravity of this post and its potential ramifications, I chose to make this post anonymously.
While the reliability of this post does suffer for the reasons you noted, I will say that I tried to largely focus this post around information originating either directly from Annie herself or from Elizabeth Weil (who had direct communication with Annie) in way that is independent of my (understandably suspicion-inducing) identity (er, lack thereof.) I wanted this post to be centered around Annie, and what she has claimed, rather than my information-limited interpretation of her and the claims she’s made.
I share your concern, not only about the reliability of Annie’s flashbacks, but also about the validity of the claims she’s made as a whole. As I note in my response to “Objection 4”, Annie has provided no direct evidence to corroborate her claims, to the best of my knowledge.
I also acknowledge that the links I provided (e.g. from saprea.org) do not meet rigorous standards that would enable me to label them as “scientific” or “empiric” evidence to corroborate Annie’s account. I provide them merely as a way of noting that the symptoms that Annie’s reported seem plausible.
As I mentioned, the intent of this post is to promote discussion about the claims that Annie has made, and to spread awareness of the fact that Sam has not yet responded to Annie’s (very serious claims.) This post does not claim that Annie’s claims are provably or indisputably valid. In fact, I think the opposite is true: her claims are not yet corroborated by direct evidence, and they certainly are disputable. I currently hold Sam Altman to be innocent, until proven guilty.
In spite of this, I still thought that this post was worth making, as a means of bringing attention to Annie’s claims, which I think have a nonzero probability of being true in whole or in part.
I think the points you make are somewhat valid. I don’t entirely agree with the reasoning from which they originate.
While I agree that:
-- Yes, it is not necessary for a person exploring their sexuality to do so by sexually assaulting a younger family member
-- Yes, providing “13-year-old Sam Altman was exploring his sexuality” as the explanatory motive of 13-year-old Sam’s sexual assault of 4-year-old Annie is not entirely satisfactory},
I do not agree that:
-- 13-year-old Sam Altman choosing to explore his sexuality by sexuality assaulting his 4-year-old sister is a psychologically infeasible (I do acknowledge that this is not exactly the claim you are making.)
I also think that Annie may not have been fully literal in her provision of “13-year-old Sam Altman was exploring his sexuality” as the explanatory motive for him sexually assaulting her.
I do acknowledge that this may not provide an entirely satisfactory explanation of why a 13-year-old Sam (purportedly) chose to sexually abuse a 4-year-old Annie Altman. Nevertheless, I do not think that {a 13-year-old Sam Altman sexually abusing a 4-year-old Annie Altman} is mutually exclusive with {Sam Altman coming out as gay as a teenager, and being openly gay since then.}
I also find Annie’s claims emotionally difficult to read. Annie’s claims are very serious. Though, as I have acknowledged, their validity has yet to be convincingly established.
I also would be interested to know what evidence led her to believe she had been widely shadowbanned. In general, I would be interested to hear more from Annie, Sam, or those close to this.
Sorry for the delayed response—yes, I think this kind of gets at the heart of the matter. I think, though I did a pretty good job with being rational in this post, and trying to make rational, unbiased claims from/using the information that exists, I could have been a bit more refined and clear-cut.
I honestly feel a bit bad, because this is an important issue, and I hope I didn’t screw things up by (unintentionally) presenting things in a irrational or biased way. I’ll try to be very rational and unbiased in this comment.
I think my statement that I was “trying to figure out the truth” in an earlier comment was misguided and imprecise. You were keen to notice this. In a situation like this, there are large amounts of uncertainty, and there is currently no proof of misconduct (that I’ve seen.)
I think what this post does is {provide a (relatively) accurate description of the state of affairs regarding Annie’s claims.} I do feel pretty good about the way in which I presented the information relevant to this matter in this post. Though I don’t want to necessarily “take shots” at Elizabeth Weil, whose nymag article provided basically the only significant written third-party acknowledgment of Annie’s claims, I will say that I prefer the (hopefully, more) objective, straight-from-the-source, uncertainty-acknowledging approach I’ve taken here.
The key thing here is that, currently, the primary information we have is:
Claims that Annie has made on social media, as well as a few pictures of her from when she was sick that she took, and a few screenshots of her social media that potentially indicate, but do not provably or definitively, indicate that she experienced shadowbanning, let alone that the low engagement/shadowbanning occured because of Sam. It is important to avoid the conjunction fallacy:
Let A = the event that Annie Altman, or (digital) media relating to her did indeed experience shadowbanning, low engagement, etc.
Let B = the event that Sam Altman caused A to occur.
Then
P(A ∧ B) ≤ P(A).
To me, it seems very hard to prove that one has been shadowbanned. To me, this would require proof of an arrangement between a “shadowbanning-requester” (e.g. Sam Altman) and the “shadowbanners” (e.g. developers or mods at Instagram, X, etc.), or direct evidence of actions taken and/or code written by mods, devs, etc. that shadowbanned Annie’s content. In this matter, that has not been provided.
A 2018 podcast episode that Annie did with Sam, Jack, and Max. Yes, while it is potentially suspicious that Sam cut Annie off around 24:50 ish, it doesn’t prove anything.
A Twitter post from 2018 where Sam Altman shared a link to Annie’s Youtube channel.
A variety of other social media posts from Annie that, while they are not inconsistent with the story she is telling / claims she is making about Sam, do not provide proof for the claims she has made about Sam.
For example, Annie seems to have posted multiple social media posts showing her in Hawai’i at the times that she claims she was. So this does corroborate the part of the larger claim-story in which Annie claims she was in Hawai’i at time X. However, these only support that individual part of her story; they provide no evidence for anything else.
So, I think the main thing that this post has going for it is that it aggregates what is out there in a relatively objective/unbiased way. That is, it aggregates (many of) the claims Annie has made, and related media that exists on the Internet.