I'm no big fan of OpenAI myself, but what he was whistleblowing over (OpenAI training on copyrighted material) is not really a secret, nor is it the kind of knowledge an organisation would kill over, even if he had evidence to present.
I hope the autopsy gives them peace, but I'm not expecting it to change the result :/
This is where I stand. I think corporations have assassinated whistleblowers in the past, maybe even the very recent past, but I don't see what they stand to gain here unless they were just sending a message to everyone else.
"sending a message to everyone else" - is a pretty big incentive.
Think about Boeing's whistleblowers.
Even if they do not kill them directly, they bury them under so much legal fees that they are financially ruined for life, their career broken and through clandestine PR operations destroy even their social life and public image to make an example of them.
I'd just point out that assassination doesn't have to be logical activity.
Mohammed bin Salman and the Saudi government were aware of the repercussions of murdering Jamal Khashoggi, if it were found out, but went ahead and did it anyway.
What they and other companies are doing with available publicly available data (whether copyrighted or not) is a matter of fact and subject to law. The opinion of their executives on copyright regimes is pretty much irrelevant.
> the kind of knowledge an organisation would kill over
That's the wrong place to look a motive because you are right, the consequences of corporate malfeasance are rarely existential for the organisation. They pay the fine, fire whoever broke the law and move on... and here we find where to look for motive.
A killing would not be ordered by a boardroom but by an individual or tight circle for their direct personal benefit rather than wider collective benefit. If an executive is looking to lose millions, personal legal consequences and a ruined career... now there is motive.
There is also the less logical corporate criminal (e.g. ebay) where it's not personal benefit but a tornado of ego, paranoia and delusion. They escalate some minor business beef into a mind-consuming psychopathic obsession taking ever more extreme actions against their perceived enemy.
Even if companies do not directly kill whistleblowers, they may use targeted strategies that bury them under so much legal fees that would financially ruin them for life, have their career shattered and, through clandestine PR operations destroy their social life and public image to make an example of them, which leaves them with nothing left to fight for.
Or the company hires publicists who specialize in character assassination.
There's a NY Times story about how this sort of thing is done around celebrities. 90% of what you read, watch, or listen to involving celebrities is manufactured attention-seeking or revenge PR.
The “hatefest” was because of the trial. There were so many inconsistencies and flawed arguments from her side, along with damaging audio recordings, that made people feel that she had betrayed their trust in her initial claims. The clips from the trial went viral repeatedly. I doubt you need manufactured PR for that.
I remember the first HN thread I saw where a lot of commenters chastised anyone who brought up the possibility of assassination, supposedly out of "respect" for the bereaved next of kin.
I wonder if the family would have found sufficient courage to do what they deem an obvious necessity without seeing others ask the same questions they apparently had themselves.
While a prediction market of actions can obviously lead to perverse incentives, a prediction market of comments may have some utility in aligning commenters utterances with their true beliefs.
Where are those commenters now?
Why aren't they chastising the parents to stop poking the hornets nest, out of respect for ... the other next of kin / friends / ...
There is no doubt he was killed.
1) For People who say everyone already knows Open AI already is infringing copyright and that is not a good enough reason for them to kill Suchir:
He was going to present documents and testify in a legal filing days before his death. Its not like Open AI was going to tell the courts ... "Ohh yeah everyone knows we steal data , its an open secret , dont you know ? " .. They obviously want to defend that.
2) He was someone very driven. Not someone sad about life sitting at home eating chips and watching tv. If he has plans to pursue further research after leaving open ai , and go after his ex-employer , he clearly is not someone beaten down by life that he would take his life. ( On one hand he is planning to present documents and testify , and on the other hands these guys are saying its a suicide ).
3) Who ransacks their own home before suicide. There were also signs of struggle, and blood spots in his apartment.
Given the vulnerability of whistleblowers, I would suggest a (poorly thought out) mitigation:
The organization being whistled at must take full responsibility for the safety of that person at least until the case is resolved.
Obviously the details of the implementation would be important, and it likely could never be completely fair, but first goal should be to remove the serious risks involved in legitimate whistleblowing.
Does this include protecting someone from suicide? How would are organization protect them from that without infringing on the whistleblower's liberty?
Yes, protection from suicides and "suicides" would be critical.
How to do that while preserving freedom is an open question. If the organization and its leaders have a genuine incentive in keeping the person safe, do you think there would be fewer deaths?
Over time, I have a feeling that a set of best practices would be developed for these situations.
Speaking as someone who has dealt professionally with several dozens of suicides and hundreds of attempts, as a paramedic, I'd like to add the following notes of caution:
It's very easy to claim that your mental health emergency was an "accident" or "overreaction", and so forth. "No, I'm not really suicidal, I just lost myself for a moment or got overwhelmed. Everything is fine now/I'm going to see someone to help me cope/get the help I need/I was never -actually- suicidal"... and then commit suicide in very short order.
Denial is a stage of grief. I have had many family members tell me to be careful when responding to their loved ones, "because it's a crime scene" or similar, "because they've never shown any mental health issues", "would never do this to themselves", "would never do it to those they loved" that... go on to be ruled suicide.
I'm not saying anything about this particular case, but it's very easy to see a smoking gun when you're predisposed to it (and I'm not someone who is a particular fan of OpenAI/SamA, etc., hardly a defender).
It's not an either-or. It could be an Aaron Swarz, the persecution by his opponents triggered his own mental health and thus a suicidal response. Here, the material effect of ostracism and isolation, destruction of his livelihood as the price of being a critic/apostate/whistleblower--that can lead people to becoming suicidal.
Never said it was, and I know nothing to understand what happened here.
I'm just saying, people should pause and reflect before jumping on the conspiracy theories immediately. It was similar with someone else recently. "If you're reading this know that I would never commit suicide". People immediately sprung to the conspiracy.
The alternative here is accepting it’s perfectly reasonable to think that Sam Altman decided it would be a good idea to have an ex-employee (of no particular import) murdered for expressing their opinion on Twitter.
have you seen the lengths to which Sama goes to silence ex-employees using legal means like cancelling vested equity and super-aggressive NDAs. there have been multiple articles written on that. its not far fetched to imagine & definitely not conspiratorial if you factor in the fact that now openAI has $40B worth of saudi investments. that regime is known to some of most vicious atrocities including that reporter.
this would be sending a clear message to future whistleblowers.
it wasnt sam. it was the federal government who has many state assets installed in openAI. would this be the first time a global power murdered someone for trying to blow the whistle?
Because it’s almost always correct and feeding oxygen to conspiracies further pollutes the space and numbs people to when there really are sinister actors at play.
I think it’s a tricky situation. Most of the time it’s noise and unproductive. But we also don’t want to sterilize the environment so far that we can’t ever question things. There’s a tricky balance.
Readit News
I hope the autopsy gives them peace, but I'm not expecting it to change the result :/
Think about Boeing's whistleblowers.
Even if they do not kill them directly, they bury them under so much legal fees that they are financially ruined for life, their career broken and through clandestine PR operations destroy even their social life and public image to make an example of them.
Mohammed bin Salman and the Saudi government were aware of the repercussions of murdering Jamal Khashoggi, if it were found out, but went ahead and did it anyway.
What they and other companies are doing with available publicly available data (whether copyrighted or not) is a matter of fact and subject to law. The opinion of their executives on copyright regimes is pretty much irrelevant.
Dead Comment
Deleted Comment
That's the wrong place to look a motive because you are right, the consequences of corporate malfeasance are rarely existential for the organisation. They pay the fine, fire whoever broke the law and move on... and here we find where to look for motive.
A killing would not be ordered by a boardroom but by an individual or tight circle for their direct personal benefit rather than wider collective benefit. If an executive is looking to lose millions, personal legal consequences and a ruined career... now there is motive.
There is also the less logical corporate criminal (e.g. ebay) where it's not personal benefit but a tornado of ego, paranoia and delusion. They escalate some minor business beef into a mind-consuming psychopathic obsession taking ever more extreme actions against their perceived enemy.
Deleted Comment
Reminds me of TV shows like Dirty Rotten Money, and Damages.
Dead Comment
Dead Comment
Dead Comment
Disclaimer: for the record, and because I think it needs to be stated directly, I don't believe this man was killed by OpenAI.
You can get away with it, but you aren't technically allowed to do it.
So what? Why is this meaningful?
There's a NY Times story about how this sort of thing is done around celebrities. 90% of what you read, watch, or listen to involving celebrities is manufactured attention-seeking or revenge PR.
https://www.nytimes.com/2024/12/26/opinion/blake-lively-just...
There seems to be a decent chance the Amber Herd hatefest on reddit and twitter was manufactured by Johnny Depp's PR team.
I wonder if the family would have found sufficient courage to do what they deem an obvious necessity without seeing others ask the same questions they apparently had themselves.
While a prediction market of actions can obviously lead to perverse incentives, a prediction market of comments may have some utility in aligning commenters utterances with their true beliefs.
Where are those commenters now?
Why aren't they chastising the parents to stop poking the hornets nest, out of respect for ... the other next of kin / friends / ...
He was going to present documents and testify in a legal filing days before his death. Its not like Open AI was going to tell the courts ... "Ohh yeah everyone knows we steal data , its an open secret , dont you know ? " .. They obviously want to defend that.
2) He was someone very driven. Not someone sad about life sitting at home eating chips and watching tv. If he has plans to pursue further research after leaving open ai , and go after his ex-employer , he clearly is not someone beaten down by life that he would take his life. ( On one hand he is planning to present documents and testify , and on the other hands these guys are saying its a suicide ).
3) Who ransacks their own home before suicide. There were also signs of struggle, and blood spots in his apartment.
Suchir Balaji's mom talks about his life, death, and disillusionment with OpenAI
https://www.businessinsider.com/suchir-balaji-openai-mom-dea...
(https://news.ycombinator.com/item?id=42516784)
Dead Comment
Obviously the details of the implementation would be important, and it likely could never be completely fair, but first goal should be to remove the serious risks involved in legitimate whistleblowing.
How to do that while preserving freedom is an open question. If the organization and its leaders have a genuine incentive in keeping the person safe, do you think there would be fewer deaths?
Over time, I have a feeling that a set of best practices would be developed for these situations.
It's very easy to claim that your mental health emergency was an "accident" or "overreaction", and so forth. "No, I'm not really suicidal, I just lost myself for a moment or got overwhelmed. Everything is fine now/I'm going to see someone to help me cope/get the help I need/I was never -actually- suicidal"... and then commit suicide in very short order.
Denial is a stage of grief. I have had many family members tell me to be careful when responding to their loved ones, "because it's a crime scene" or similar, "because they've never shown any mental health issues", "would never do this to themselves", "would never do it to those they loved" that... go on to be ruled suicide.
I'm not saying anything about this particular case, but it's very easy to see a smoking gun when you're predisposed to it (and I'm not someone who is a particular fan of OpenAI/SamA, etc., hardly a defender).
I'm just saying, people should pause and reflect before jumping on the conspiracy theories immediately. It was similar with someone else recently. "If you're reading this know that I would never commit suicide". People immediately sprung to the conspiracy.
But I don't presume to know what happened here.
Deleted Comment
What is more plausible is a "Will no one rid me of this turbulent priest?" [1] situation.
[1] https://en.wikipedia.org/wiki/Will_no_one_rid_me_of_this_tur...
this would be sending a clear message to future whistleblowers.
Personally I do not want to start any conspiracies myself by listing them, but this is hardly a binary choice
In a sense you are agreeing that this as tragic as it is probable needs no further discussion
Most here on HN seem to have fairly high critical thinking ability.
https://en.wikipedia.org/wiki/Tron_(hacker)
Now everyone is accusing others of "conspiracy theories" and browbeating everyone to follow the party line.
I think it’s a tricky situation. Most of the time it’s noise and unproductive. But we also don’t want to sterilize the environment so far that we can’t ever question things. There’s a tricky balance.
Dead Comment
Dead Comment