Says it took them 40 seconds to determine suicide. I'm not sure how that determination could be made without knowledge of the firearm's ownership. Did he legally own the firearm? If not, how did he acquire the firearm?
Is anyone else noticing the number of whistleblower deaths occurring? Boeing has had several in the past year.
The victim's grieving mother says that it took 40 seconds, with no explanation of where that number came from. I strongly suspect that this is not how the medical examiner would characterize things. Even if the police have incontrovertible evidence of what happened, they would not be running around telling the media about it for the sake of OpenAI's reputation.
I would assume if you're an overworked cop in an underfunded police department, it's gonna be hard to prove a murder from what looks like a suicide.
A toxicology is going to cost money and maybe not actually provide any more details. Any needle marks are likely to go unnoticed by the coroner doing the post mortem, if one is even given.
And the sad thing is, after all the investigation has happened, it's possible he killed himself. I'm sure lots of parents have thought their children would have never be capable of doing that, only to find out later that they did.
Someone notable, unambiguously murdered in broad daylight, with ample surveillance footage to go off of, is far easier to justify spending investigative resources on than an apparent suicide maybe being a murder. As the saying goes: "when you hear hoofbeats, think horses not zebras"
You’re just reiterating the parent commenters point, the first reason you listed was that he was “notable”. It would not be hard to find cases of unambiguous murder that go unsolved or under-investigated due to lack of notoriety.
>You’re just reiterating the parent commenters point, the first reason you listed was that he was “notable”.
The parent commenter did not mention anything about being "notable". At best he was trying to imply it by burying it in between the lines. I stated it explicitly.
Yeremi Colino, 17, stabbed to death in a fight between two street gangs. The police have published photos of the murderers but evidently nobody in that community wants to say anything about who they are. NYPD claims they are putting resources into catching them, but how was the CEO killer caught? Some random member of the public turned him in.
Which is just another way of saying that the police will not protect informants or regular citizens from many kinds of crime. The police say they're underfunded and cannot do so. Which makes this a policy decision.
How do you really expect the police to do so? Even the secret service can't stop presidential candidates from getting shot at, and you can't possibly expect the police to provide anything approaching that level of security.
That's NYC. Interesting thing about NYC is that they have a lot of cops. I mean a lot. $11B spent on policing and the police to resident ratio is 1:162. Compare that with Boise, ID say where residents pay half per person as much but the police to resident ratio is 1:563.
San Francisco exhaustively investigates when people are murdered on the street, no matter who the victim is. They have a 94% clearance rate for homicides.
Why doesn't the US gov investigate and prosecute all the people on Epstein's video tapes found in his New York mansion? When you give me the answer to that you will have the answer to your question. You are asking the right questions Obi One.
No, it is not. And their budget is similar per citizen to most major cities [1], so implying that somehow they should spend money on every whim isn’t justified.
The point is if that billion is already earmarked for normal police activities, and they're already underfunded, they can't be expected to cough up a bunch more money for this investigation.
Yes, they are so vastly underfunded they cannot follow all such wild fantasy chases without ignoring things for which they have evidence, which would necessarily put other things behind, and then you could rant about those things.
Is this answer enough for something you could simply try to understand directly instead of repeating such simplistic and myopic beliefs?
It just looks like they have a lot of money is all. People keep saying they're under funded but then link to a page that says they have almost a billion dollars per year. Much more than any other civic institution I can find info about. How much funding would it take for them to pursue suspicious deaths? What else are they doing that they don't have resources for this? Maybe the problem isn't funding.
To clarify, I'm not suggesting OpenAI was involved, or that the cause of death is anything other than suicide. I was pointing out an investigative detail (legal ownership of the weapon). I have now read 4 or 5 articles on the matter and none have mentioned this detail.
If he did legally own the weapon, it would be difficult to prove suicide isn't the cause of death without additional evidence of another person being present.
If he didn't obtain the weapon legally, that warrants an additional investigation into how he obtained the weapon (his parents' request). While it doesn't eliminate suicide as a cause of death, it is a key detail in the case.
While I doubt OpenAI is responsible for wrongdoing. OpenAI's strategic importance to AI development warrants an FBI investigation as a standard procedure.
The whistleblower may have been approached by foreign governments seeking classified information, but declined to share it. There are many potential scenarios to consider, beyond the obvious.
Very wealthy people have access to very exclusive and potentially unethical and/or illegal capabilities. Many do not think they are bound to the same rules as the rest of society. Some are probably not.
OpenAI the company may not be responsible, but there are elites in the OpenAI sphere who might be.
Both, we had seen competitors infiltration, spying attempt from foreign agents.
Back in 2008 Lead System Administrator of NTT Verios got contacted by Chinese agents and they tried to bribe him to inject html pages of some websites with browser zero day exploit. He played along and then exposed that on board meeting. That got alerted to authority and one board member who have deep roots with Chinese government booted out. Japanese people place honor above money in many cases and hard to corrupt. It won't be same in many other places.
I do not understand why SKIF's aren't being used in any kind of "lawsuit" around LLM training data.
We know that there are random undetected "landmines" buried in LLM training data. All the evidence I need is that stupid discord airforce officer doing 20 years in jail for leaking classified docs or the many stories of War-Thunder forums having classified info get leaked just so nerds can win an argument online.
I'm old man (or rather, langley's favorite useful idiot) yelling at cloud about this.
Yes, this is a plausible scenario. High-exposure tech industries tend to attract all sorts of government entities from across the world. I worked on a high exposure crypto project and witnessed everything from coercion to sabotage. You just notice people acting in weird ways and it's not clear what their agenda is.f
Interesting to see how quick people jump to conclusions. Would it not be just as reasonable that this individual was struggling with mental health issues?
I would think a whistleblower regardless of the legitimacy of claims, would be under extra stress and perhaps even likelier that folks with preexisting mental health conditions would be more inclined to go down this route. Higher chance of suicide seems more plausible than companies whacking people. It should certainly be investigated still.
Why would you say: people with preexisting mental health conditions are more likely to be whistleblowers? That's a very prejudiced statement if I'm correct in what you mean.
I'd rather say these people are more sensitive and aware of the consequences of their actions which is why they chose the path of a whistleblower, which is very different from saying they had mental health issues from the beginning.
Other way around, I did a poor job of phrasing it. There are many many whistleblowers annually . We hear about one or two of the high profile ones and of those there may be a suicide. I find suicide less surprising statistically than a company whacking the person. Especially in this case, they whacked him because he is going to tell the world they trained on copyright material? The simpler explanation is the position he was in probably led to a mental health crisis. I imagine a lot of his coworkers were friends, gone, company gone. Lots of potential stress for a younger person.
Plenty of mental health conditions make a person more likely to have a stronger reaction to a situation, to take risks, or to act without consideration of consequences. Following up a whistleblower act with suicide because it didn’t go as hoped is also not surprising.
I think many people underestimate the mental anguish caused by being ostracized by your former coworkers, and by the realization that, even after a lifetime of working hard, your career prospects are now in tatters.
The new method to manage political dissidents/whistleblowers in the 21st century is chemical weapons that produce profound states of depression/anxiety that can bring about obsessive thoughts about suicide.
These poisons work miracles. If the person talks about it, most people will believe they are crazy before believing they are being poisoned. There is no way to prove you have been poisoned without spending millions of dollars. All of the advances in medical science pave the way for these poisons to be developed and manufactured cheaply.
OpenAI has direct connections to the military industrial complex in the US. The leaders in these orgs have huge egos that don't like to be double-crossed, and despite it not making sense long term for their business, they are perfectly safe to retaliate using these methods and choose to do so.
> These poisons work miracles. If the person talks about it, most people will believe they are crazy before believing they are being poisoned. There is no way to prove you have been poisoned without spending millions of dollars.
They’re called satellites. They probably made him hallucinate.
> The leaders in these orgs have huge egos that don't like to be double-crossed, and despite it not making sense long term for their business, they are perfectly safe to retaliate using these methods and choose to do so.
When billions - perhaps trillions - of dollars are on the line, CEOs and their associates will stop at nothing to get that money. They're surely not going to let the life of a single person stand in their way.
Look at what the eBay CEO did to someone simply criticizing them.
>Look at what the eBay CEO did to someone simply criticizing them.
If you're trying to imply openai killed him and that's the best example of corporate retaliation you can come up with, then you're nowhere close to proving your point. Sending pigs heads is despicable, by nowhere close to ordering a hit.
I listened to the podcast "Kill List" where journalists got backend access to a hitman for hire website and do episodes featuring targeted individuals they notified.
What I got out of it is a garden variety abuser crosses the threshold to actually ordering a hit when the abuser felt they lost power over the abused.
A whistleblower about to publicly testify would 100% match up with that.
Again, that's the claim you're trying to argue for, but the best evidence you presented so far (ebay harassment case) doesn't prove that. It shows that they're at least not willing to commit murder. Maybe there's some psycho CEO out there that did order a hit, but you haven't presented that, and it's a stretch to go from "sending pig's head" to "ordering a hit".
I'm implying that is something that needs to be investigated. The motive is there: hundreds of billions of dollars. And people do not become billionaires by being nice and playing nice. Many of them have clinical psychopathy (ASPD).
Do you think the Boeing whistle blowers just keep running into bad luck?
>Do you think the Boeing whistle blowers just keep running into bad luck?
"Bad luck" implies they're dying at a higher rate than expected. Do you have any evidence that's the case, factoring in the fact that being a whistleblower is stressful? Stunt drivers probably die in occupational related deaths at a far higher rate than expected, but I don't think it's "bad luck" or make posts calling for police to reinvestigate on the off chance that the accident was actually a murder.
Excerpt: "They knew armed guards would be required to protect their compounds from raiders as well as angry mobs. One had already secured a dozen Navy Seals to make their way to his compound if he gave them the right cue. But how would he pay the guards once even his crypto was worthless? What would stop the guards from eventually choosing their own leader? The billionaires considered using special combination locks on the food supply that only they knew. Or making guards wear disciplinary collars of some kind in return for their survival."
"Seven eBay employees pleaded guilty to charges involving criminal conspiracies.[3][4] The seven employees included two senior members of eBay’s corporate security team.[5] Two members of eBay's Executive Leadership Team who were implicated in the scandal were not charged"
The last paragraph of the article hit me very hard:
“I don’t know how I could have saved my son by teaching him to tell lies,” Ramarao said. “The ethics with which I raised him took his life today.”
Even calling him a whistleblower is quite a stretch, we already know companies like Open AI are based on stealing copyrighted work from creative humans.
No. We knew this already because it's evident in the output. For example, when MidJourney can perfectly replicate the visual style of a specific artist whose work is copyrighted.
Could you point out the text in copyright law that entitles anyone to ownership of a visual style?
Nor have I ever heard of a hardware or software tool ever being responsible for the output of the end user.
If I use a copy machine to make a duplicate of your work, is Xerox responsible? I copy your clip art in Illustrator, is Adobe the infringing party? If I make a pencil copy, are we to name Ticonderoga as a defendant? What if I coaxed the entirety of your copyrighted work out of the word suggestion engine on my phone’s keyboard?
I don't understand your point. The whistleblower didn't "prove" it either.
There is already enough publicly-visible info for copyright holders to sue OpenAI even without the whistleblower. And when a judge allows the civil suit to proceed then the internal documents will come out during discovery, even without the whistleblower.
So, yeah the whistleblower lends more weight to the argument, but there was already enough to go on even without him.
Well, someone committed suicide and they also went against the company by challenging them publicly. A company that makes a truckload of money. Shouldn't an investigation be in order?
I did a quick sketch of premises, no warranty :) Curious for your thoughts, tl;dr: imho there doesn't seem to be a reason to investigate OpenAI's working conditions.
- Person A died
- Person A used to work for company X
- Person A spoke negatively about company X's choice in training data sourcing
- Company X has a lot of money
- Law enforcement should investigate Company X workplace conditions
I think the missing piece is some sort of evidence Company X was involved, or finding some tangible evidence that someone who committed suicide was materially, and thus criminally, caused by workplace conditions at a job they left a while ago. (this is a pretty high bar, a civil suit has ~0 bar and would be easier, but law enforcement isn't responsible for helping with those)
The public funds uninterested cross-discipline teams of highly trained professionals to crosscheck eachother on ex. cause of death, and without any evidence from that process indicating a possiblity someone other than the decedant was involved...it appears to be a leap.
Not true. Only the highest level researchers get paid a ludicrous amount. Most engineers at OpenAI in SF would get paid about 300k which can barely afford the taxes and rent in SF.
Is that really true? I saw some one-bedroom apartments there for $3500 -- expensive but even with that, 300K is a decent amoount. According to an online tax calculator, the take home pay is 185,000 or 15.4K per month.
Even if you spend a bit and spend another 3K per month on stuff like food, bills, and some spending money, that's still almost 9K that you can save per month. Sounds VERY lavish to me. So lavish that in 15 years of work you could save up 1.6 million, buy a cheap house somewhere (or already have one from buying it while working there) for $500K and have 1.1 million to live frugally on and just work on side projects for a little extra cash.
The strange thing about this retirement plan is that it assumes you will spend 15 years living somewhere, and then one day just walk away from all the social ties you've made during that period, during the prime of your life. I suppose there are people who could probably do this, but for most folks, this is not going to be a joyful path.
Wildly incorrect. $300k will let you live comfortably in SF, without roommates.
The fact that some people cannot live alone on that salary in SF does not mean most people cannot. I make considerably less than that and I could make it work. I wouldn't, because that would destroy my ability to save anything, but I could do it.
But not for the Bay Area. As of November 2024, the median price of a home in the Bay Area is around $1,316,500.
On a 3:1 debt to income ratio, I'd put median income for the area for those who can afford to buy a home there is approx $350K (80% financing on median home price).
The idea that OpenAI had him killed is beyond ludicrous. Even if it were in any way realistic that OpenAI leadership somehow had access to hit men (!!), killing this guy would have been absolutely pointless as far as OpenAI's possible legal troubles go. The notion is not only outlandish, but also illogical.
He publicly stated OpenAI used copyright materials. What would killing him now possibly accomplish? Is there some legal statute I'm unaware of where a company can't be investigated if the original whistleblower is dead? It's not like he's the sole witness to a crime, and without his testimony the case disappears.
The Steiners were harassed and threatened both online and physically in their home by deliveries of such things as a bloody pig mask, live cockroaches and spiders, a funeral wreath, and large orders of pizza.[5][1][6] Pornographic magazines with David Steiner’s name on them were sent to a neighbor’s house.[5][1][6]
Employees flew from California to Boston so they could vandalize the couple's Natick, Massachusetts home as well as stalk their personal vehicle.[1][6] Plans were even made to break into the couple's garage and place a GPS tracker on their car.[1]
Having access to hit men isn't actually that difficult, the hard part is telling which ones are undercover police.
Murderers don't act rationally. If someone at OpenAI did kill this guy, it wouldn't have been because they thought it was in their interest. It would have been because they had some kind of emotional problem or an extreme case of entitlement and thought they could get away with it. People in all walks of life will commit crimes, and guessing innocence on the basis of rational motives is about as reliable as guessing guilt.
It's absolutely bonkers to think that the leadership of OpenAI thought to themselves: "This employee who spilled the beans on our use of copyrighted material, we better off him before he says more." That's just nuts. It's the weakest-sauce of possible things to blow a whistle on anyway, given how content owners are already pitchforks-out against AI companies since day one. He's a random employee who came out and said "That obvious thing you all think OpenAI is doing (slurping in copyrighted data)... yup they did that." And exactly how woult killing him would fix their problem anyway?
It doesn’t have to be the CEO or the execs directly and there are many plausible stories that could have occurred.
What if it was a former coworker who is deeply in gambling debt and needs OpenAI to IPO in order for him to not lose everything in his life?
I think the people around the whistleblower would know better whether or not he was in the mental state to commit suicide. No note for his parents is pretty sus, especially given all the plans they made to visit.
I can only imagine you being a "1st world" citizen for being this naive.
> "OpenAI somehow had access to hit men (!!)"
You think hitmen are only for movies?? Even much much smaller scale companies have people who sort out their problems. OpenAI is one of the biggest companies. Of course it has hitmen, connections with the police, FBI, CIA, NSA and whatever you can think of.
These companies (Boeing, OpenAI, Google, Microsoft) are sending a MESSAGE to the future whistleblowers. "You step on our tail, you're done" and they are making it as obvious as possible and you idiots still don't get the message?
You think these are some cute goofy nerds sitting and doing tech stuff and playing warcraft all day?
Testifying during a trial? I am not saying this happened, but it is not unheard of witnesses in high-profile trials to get murdered (and getting covered up) even if they have been public about it. Getting public about it means nothing in terms of testimonies during the trial, and if the person is dead there is no witness's testimony to be used.
I do not know about the specifics in "open"AI's court cases, but the fact if him being public in general does not mean much, as I understand it. Moreover, it can scare future whistleblowers. The motive is definitely there, now if this is what happened is a different story.
> "The idea that OpenAI had him killed is beyond ludicrous."
On the other hand, given that everyone either assumes the whistle-blower was murdered, or at least that the stress of the situation contributed to the suicide, why did OpenAI not do absolutely everything in their power to keep the person safe and mentally well?
Why should OpenAI have zero liability for the death, which anyone could have foreseen was a reasonably possible outcome?
It just seems like whistleblower protection laws could be much more effective if the company had genuine incentive in the person not dying.
I don't have any skin in the game as far as this particular case goes but if there was a list of CEOs who would order a hit, Sam Altman would 110% be near the top of that list. The guy is a straight-up sociopath.
> What would killing him now possibly accomplish?
Dissuading future potential whistleblowers of course.
He always makes very loud, bold statements to make himself look good and then quietly reverses them later on.
I don’t care about money.
But he drives a $4 million car.
OpenAI will never be for profit.
Actually we will create a for-profit wing.
I will never take equity.
My investors want me to take an equity stake, not me!
OpenAI will not be used for weapons.
OpenAI will be used for defensive weapons only.
Don’t believe anything he says at this point. He wants AGI and he wants to cut everyone else out and he wants to be in control of it and become the next richest and most powerful man in the world.
as is their right. Companies should be held responsible for at least putting undue social, financial and legal pressure on people who are speak out for their beliefs regarding public matters.
Does the family have a concrete basis for suspicion of such? The police don't have any particular power to investigate wild hypotheses. They can ask his friends and coworkers "hey did you kill him", and they'll respond with some combination of "no" and "talk to my lawyer" - then what?
Total coincidence, nothing to see here, peasants. If you think otherwise you’re a conspiracy theorist, a Putin apologist, a gaslighter, or whatever we’re calling you today.
What the hell is wrong with peoples brains?
Everyone in these comments are like "Oh not OpenAI wouldn't do that", "maybe he had mental health issues" Shut up please. How naive are you people?
Magically every whistleblower is dying and they all have mental health issues? God must hate whistleblowers then.
It's easier to believe companies are your friends than to consider that the elites would do anything to save their money.
Americans and Western Europeans are in general so damn naive when it comes up to that. Like sure, live in your fantasy where everyone abides the law and no bad actors exist :<
The amount of harassment and hate being propagated by the media certainly didn't help. I would also find The New York Times to be liable for all the AI doomerism they propagate.
Says it took them 40 seconds to determine suicide. I'm not sure how that determination could be made without knowledge of the firearm's ownership. Did he legally own the firearm? If not, how did he acquire the firearm?
Is anyone else noticing the number of whistleblower deaths occurring? Boeing has had several in the past year.
https://fortune.com/2024/05/02/boeing-whistleblower-dead-jos...
The victim's grieving mother says that it took 40 seconds, with no explanation of where that number came from. I strongly suspect that this is not how the medical examiner would characterize things. Even if the police have incontrovertible evidence of what happened, they would not be running around telling the media about it for the sake of OpenAI's reputation.
I would assume if you're an overworked cop in an underfunded police department, it's gonna be hard to prove a murder from what looks like a suicide.
A toxicology is going to cost money and maybe not actually provide any more details. Any needle marks are likely to go unnoticed by the coroner doing the post mortem, if one is even given.
And the sad thing is, after all the investigation has happened, it's possible he killed himself. I'm sure lots of parents have thought their children would have never be capable of doing that, only to find out later that they did.
But when a different citizen (CEO) dies, they have the resources for a hundred detectice manhunt
Someone notable, unambiguously murdered in broad daylight, with ample surveillance footage to go off of, is far easier to justify spending investigative resources on than an apparent suicide maybe being a murder. As the saying goes: "when you hear hoofbeats, think horses not zebras"
You’re just reiterating the parent commenters point, the first reason you listed was that he was “notable”. It would not be hard to find cases of unambiguous murder that go unsolved or under-investigated due to lack of notoriety.
>You’re just reiterating the parent commenters point, the first reason you listed was that he was “notable”.
The parent commenter did not mention anything about being "notable". At best he was trying to imply it by burying it in between the lines. I stated it explicitly.
There was a brown kid stabbed to death in NYC the same day. No resources were expended to find his killer.
Yeremi Colino, 17, stabbed to death in a fight between two street gangs. The police have published photos of the murderers but evidently nobody in that community wants to say anything about who they are. NYPD claims they are putting resources into catching them, but how was the CEO killer caught? Some random member of the public turned him in.
Yes and unfortunately there's a saying that goes something like snitches get stitches. Talking to the cops is the last thing many people want to do.
Which is just another way of saying that the police will not protect informants or regular citizens from many kinds of crime. The police say they're underfunded and cannot do so. Which makes this a policy decision.
[1] https://www.polfed.org/news/blogs/2024/bbc-panorama-highligh...
[2] https://www.theatlantic.com/ideas/archive/2023/03/underfundi...
(trying to keep any partisan links out of it, as there are MANY blaming one side or the other)
How do you really expect the police to do so? Even the secret service can't stop presidential candidates from getting shot at, and you can't possibly expect the police to provide anything approaching that level of security.
What about the migrant killed? NYT didn’t even report it.
In one of the busiest square .5 miles on the planet, no less.
That's NYC. Interesting thing about NYC is that they have a lot of cops. I mean a lot. $11B spent on policing and the police to resident ratio is 1:162. Compare that with Boise, ID say where residents pay half per person as much but the police to resident ratio is 1:563.
[flagged]
San Francisco exhaustively investigates when people are murdered on the street, no matter who the victim is. They have a 94% clearance rate for homicides.
94% feels low when they are determining suicide in only 40s
Why doesn't the US gov investigate and prosecute all the people on Epstein's video tapes found in his New York mansion? When you give me the answer to that you will have the answer to your question. You are asking the right questions Obi One.
Welcome to the USA buddy.
> I would assume if you're an overworked cop in an underfunded police department, it's gonna be hard to prove a murder from what looks like a suicide.
In some countries, a doctor clarifies the cause of death.
* In almost all countries
I don't know any developed country where a trained coroner does not do handle unnatural deaths. But apparently USA likes to do its own thing.
Isn't their budget literally like a billion dollars a year?
No, it is not. And their budget is similar per citizen to most major cities [1], so implying that somehow they should spend money on every whim isn’t justified.
[1] https://www.vera.org/publications/what-policing-costs-in-ame...
Well, and are they underfunded or not?
The point is if that billion is already earmarked for normal police activities, and they're already underfunded, they can't be expected to cough up a bunch more money for this investigation.
But are they underfunded?
Yes, they are so vastly underfunded they cannot follow all such wild fantasy chases without ignoring things for which they have evidence, which would necessarily put other things behind, and then you could rant about those things.
Is this answer enough for something you could simply try to understand directly instead of repeating such simplistic and myopic beliefs?
It just looks like they have a lot of money is all. People keep saying they're under funded but then link to a page that says they have almost a billion dollars per year. Much more than any other civic institution I can find info about. How much funding would it take for them to pursue suspicious deaths? What else are they doing that they don't have resources for this? Maybe the problem isn't funding.
[flagged]
Another theory is that whistleblowers are under a lot of stress. Pre-existing emotional issues would be exacerbated, sometimes leading to suicide.
To clarify, I'm not suggesting OpenAI was involved, or that the cause of death is anything other than suicide. I was pointing out an investigative detail (legal ownership of the weapon). I have now read 4 or 5 articles on the matter and none have mentioned this detail.
If he did legally own the weapon, it would be difficult to prove suicide isn't the cause of death without additional evidence of another person being present.
If he didn't obtain the weapon legally, that warrants an additional investigation into how he obtained the weapon (his parents' request). While it doesn't eliminate suicide as a cause of death, it is a key detail in the case.
While I doubt OpenAI is responsible for wrongdoing. OpenAI's strategic importance to AI development warrants an FBI investigation as a standard procedure.
The whistleblower may have been approached by foreign governments seeking classified information, but declined to share it. There are many potential scenarios to consider, beyond the obvious.
Very wealthy people have access to very exclusive and potentially unethical and/or illegal capabilities. Many do not think they are bound to the same rules as the rest of society. Some are probably not.
OpenAI the company may not be responsible, but there are elites in the OpenAI sphere who might be.
Yes, when I am was a system engineer in NTT DOCOMO, we have seen cases of such attempt by foreign nations. It was more common than what you think.
It might not have been a foreign nation (?).
Both, we had seen competitors infiltration, spying attempt from foreign agents. Back in 2008 Lead System Administrator of NTT Verios got contacted by Chinese agents and they tried to bribe him to inject html pages of some websites with browser zero day exploit. He played along and then exposed that on board meeting. That got alerted to authority and one board member who have deep roots with Chinese government booted out. Japanese people place honor above money in many cases and hard to corrupt. It won't be same in many other places.
Everyone doing anything interesting in AI research should assume that they are targets of this kind of thing.
> Everyone doing anything interesting in AI research
They were just using copyrighted material for training. I'm surprised that, OpenAI being a Microsoft child, BSA kept quiet.
I do not understand why SKIF's aren't being used in any kind of "lawsuit" around LLM training data.
We know that there are random undetected "landmines" buried in LLM training data. All the evidence I need is that stupid discord airforce officer doing 20 years in jail for leaking classified docs or the many stories of War-Thunder forums having classified info get leaked just so nerds can win an argument online.
I'm old man (or rather, langley's favorite useful idiot) yelling at cloud about this.
Yes, this is a plausible scenario. High-exposure tech industries tend to attract all sorts of government entities from across the world. I worked on a high exposure crypto project and witnessed everything from coercion to sabotage. You just notice people acting in weird ways and it's not clear what their agenda is.f
> High-exposure tech industries tend to attract all sorts of government entities from across the world.
Yeah this is correct. I’ve experienced the same thing working with crypto folks.
Interesting to see how quick people jump to conclusions. Would it not be just as reasonable that this individual was struggling with mental health issues?
I would think a whistleblower regardless of the legitimacy of claims, would be under extra stress and perhaps even likelier that folks with preexisting mental health conditions would be more inclined to go down this route. Higher chance of suicide seems more plausible than companies whacking people. It should certainly be investigated still.
Why would you say: people with preexisting mental health conditions are more likely to be whistleblowers? That's a very prejudiced statement if I'm correct in what you mean.
I'd rather say these people are more sensitive and aware of the consequences of their actions which is why they chose the path of a whistleblower, which is very different from saying they had mental health issues from the beginning.
Other way around, I did a poor job of phrasing it. There are many many whistleblowers annually . We hear about one or two of the high profile ones and of those there may be a suicide. I find suicide less surprising statistically than a company whacking the person. Especially in this case, they whacked him because he is going to tell the world they trained on copyright material? The simpler explanation is the position he was in probably led to a mental health crisis. I imagine a lot of his coworkers were friends, gone, company gone. Lots of potential stress for a younger person.
> they whacked him because he is going to tell the world they trained on copyright material?
"Smithers, have this young man killed immediately before he talks to the press!"
"But sir, he's already gone public."
"Oh no, our secret is out!"
"No sir, the thing he revealed was already widely known."
"Then we must get rid of him to stop any civil suit."
"No sir, killing him actually won't stop any civil suit at this point."
"Then get to it post haste! Find me the first-available hit man on X. And keep this quiet."
Plenty of mental health conditions make a person more likely to have a stronger reaction to a situation, to take risks, or to act without consideration of consequences. Following up a whistleblower act with suicide because it didn’t go as hoped is also not surprising.
Perhaps you’ve never been around bipolar folks.
Calling it “prejudiced” is silly.
That's not what I read from infecto's post, they say that people with preexisting mental health issues are more likley to commit suicide.
[dead]
I think many people underestimate the mental anguish caused by being ostracized by your former coworkers, and by the realization that, even after a lifetime of working hard, your career prospects are now in tatters.
The new method to manage political dissidents/whistleblowers in the 21st century is chemical weapons that produce profound states of depression/anxiety that can bring about obsessive thoughts about suicide.
These poisons work miracles. If the person talks about it, most people will believe they are crazy before believing they are being poisoned. There is no way to prove you have been poisoned without spending millions of dollars. All of the advances in medical science pave the way for these poisons to be developed and manufactured cheaply.
OpenAI has direct connections to the military industrial complex in the US. The leaders in these orgs have huge egos that don't like to be double-crossed, and despite it not making sense long term for their business, they are perfectly safe to retaliate using these methods and choose to do so.
Ludicrous. Extraordinary claims require extraordinary evidence, etc.
> These poisons work miracles. If the person talks about it, most people will believe they are crazy before believing they are being poisoned. There is no way to prove you have been poisoned without spending millions of dollars.
They’re called satellites. They probably made him hallucinate.
> The leaders in these orgs have huge egos that don't like to be double-crossed, and despite it not making sense long term for their business, they are perfectly safe to retaliate using these methods and choose to do so.
I’m pretty sure the guys at OpenAI don’t care.
When billions - perhaps trillions - of dollars are on the line, CEOs and their associates will stop at nothing to get that money. They're surely not going to let the life of a single person stand in their way.
Look at what the eBay CEO did to someone simply criticizing them.
>Look at what the eBay CEO did to someone simply criticizing them.
If you're trying to imply openai killed him and that's the best example of corporate retaliation you can come up with, then you're nowhere close to proving your point. Sending pigs heads is despicable, by nowhere close to ordering a hit.
https://en.wikipedia.org/wiki/EBay_stalking_scandal
I listened to the podcast "Kill List" where journalists got backend access to a hitman for hire website and do episodes featuring targeted individuals they notified.
What I got out of it is a garden variety abuser crosses the threshold to actually ordering a hit when the abuser felt they lost power over the abused.
A whistleblower about to publicly testify would 100% match up with that.
It represents the same psychology on the part of the perpetrator.
...that the (seemingly) most insane executive we know of is sane enough to not commit murder?
The people who commit these crimes are not bounded between ranges like normal people.
Again, that's the claim you're trying to argue for, but the best evidence you presented so far (ebay harassment case) doesn't prove that. It shows that they're at least not willing to commit murder. Maybe there's some psycho CEO out there that did order a hit, but you haven't presented that, and it's a stretch to go from "sending pig's head" to "ordering a hit".
Your argument is a catch 22.
I'm implying that is something that needs to be investigated. The motive is there: hundreds of billions of dollars. And people do not become billionaires by being nice and playing nice. Many of them have clinical psychopathy (ASPD).
Do you think the Boeing whistle blowers just keep running into bad luck?
>Do you think the Boeing whistle blowers just keep running into bad luck?
"Bad luck" implies they're dying at a higher rate than expected. Do you have any evidence that's the case, factoring in the fact that being a whistleblower is stressful? Stunt drivers probably die in occupational related deaths at a far higher rate than expected, but I don't think it's "bad luck" or make posts calling for police to reinvestigate on the off chance that the accident was actually a murder.
Here's an article that sheds light on how many of these billionaires think: https://www.theguardian.com/news/2022/sep/04/super-rich-prep...
Excerpt: "They knew armed guards would be required to protect their compounds from raiders as well as angry mobs. One had already secured a dozen Navy Seals to make their way to his compound if he gave them the right cue. But how would he pay the guards once even his crypto was worthless? What would stop the guards from eventually choosing their own leader? The billionaires considered using special combination locks on the food supply that only they knew. Or making guards wear disciplinary collars of some kind in return for their survival."
Your argument is that owning a doomsday bunker means you're the type of person who'd order hits?
> I'm implying that is something that needs to be investigated
See the link to Ebay stalking. Executives are above the law.
"Seven eBay employees pleaded guilty to charges involving criminal conspiracies.[3][4] The seven employees included two senior members of eBay’s corporate security team.[5] Two members of eBay's Executive Leadership Team who were implicated in the scandal were not charged"
Speaks for itself.
https://news.ycombinator.com/item?id=42524190
The last paragraph of the article hit me very hard:
“I don’t know how I could have saved my son by teaching him to tell lies,” Ramarao said. “The ethics with which I raised him took his life today.”
Even calling him a whistleblower is quite a stretch, we already know companies like Open AI are based on stealing copyrighted work from creative humans.
> we already know companies like Open AI are based on stealing copyrighted work from creative humans
No court or law says it's stealing.
> we already know companies like
and we know this because...whistleblowers???
No. We knew this already because it's evident in the output. For example, when MidJourney can perfectly replicate the visual style of a specific artist whose work is copyrighted.
Could you point out the text in copyright law that entitles anyone to ownership of a visual style?
Nor have I ever heard of a hardware or software tool ever being responsible for the output of the end user.
If I use a copy machine to make a duplicate of your work, is Xerox responsible? I copy your clip art in Illustrator, is Adobe the infringing party? If I make a pencil copy, are we to name Ticonderoga as a defendant? What if I coaxed the entirety of your copyrighted work out of the word suggestion engine on my phone’s keyboard?
No, you could guess, surmise, postulate, theorize, but until it was proven, it wasn't known
I don't understand your point. The whistleblower didn't "prove" it either.
There is already enough publicly-visible info for copyright holders to sue OpenAI even without the whistleblower. And when a judge allows the civil suit to proceed then the internal documents will come out during discovery, even without the whistleblower.
So, yeah the whistleblower lends more weight to the argument, but there was already enough to go on even without him.
I also wouldn't trust anything related to big tech which happens in California. They basically own the place.
I'm generally afraid of the US these days.
I hope they seriously investigate OpenAI and their working conditions.
What's wrong with OpenAI working conditions?
Well, someone committed suicide and they also went against the company by challenging them publicly. A company that makes a truckload of money. Shouldn't an investigation be in order?
I did a quick sketch of premises, no warranty :) Curious for your thoughts, tl;dr: imho there doesn't seem to be a reason to investigate OpenAI's working conditions.
- Person A died
- Person A used to work for company X
- Person A spoke negatively about company X's choice in training data sourcing
- Company X has a lot of money
- Law enforcement should investigate Company X workplace conditions
I think the missing piece is some sort of evidence Company X was involved, or finding some tangible evidence that someone who committed suicide was materially, and thus criminally, caused by workplace conditions at a job they left a while ago. (this is a pretty high bar, a civil suit has ~0 bar and would be easier, but law enforcement isn't responsible for helping with those)
The public funds uninterested cross-discipline teams of highly trained professionals to crosscheck eachother on ex. cause of death, and without any evidence from that process indicating a possiblity someone other than the decedant was involved...it appears to be a leap.
they get paid a ludicrous amount. the status of their working conditions are inconsequential.
No, it’s really not.
Not true. Only the highest level researchers get paid a ludicrous amount. Most engineers at OpenAI in SF would get paid about 300k which can barely afford the taxes and rent in SF.
Is that really true? I saw some one-bedroom apartments there for $3500 -- expensive but even with that, 300K is a decent amoount. According to an online tax calculator, the take home pay is 185,000 or 15.4K per month.
Even if you spend a bit and spend another 3K per month on stuff like food, bills, and some spending money, that's still almost 9K that you can save per month. Sounds VERY lavish to me. So lavish that in 15 years of work you could save up 1.6 million, buy a cheap house somewhere (or already have one from buying it while working there) for $500K and have 1.1 million to live frugally on and just work on side projects for a little extra cash.
The strange thing about this retirement plan is that it assumes you will spend 15 years living somewhere, and then one day just walk away from all the social ties you've made during that period, during the prime of your life. I suppose there are people who could probably do this, but for most folks, this is not going to be a joyful path.
Well, the point was not the plan, but the ludicrous amount of money you still get and could still use to live there if you wish.
300k is a lot of money. You can afford to live in SF quite comfortably on that. And save to buy a house and raise kids.
Wildly incorrect. $300k will let you live comfortably in SF, without roommates.
The fact that some people cannot live alone on that salary in SF does not mean most people cannot. I make considerably less than that and I could make it work. I wouldn't, because that would destroy my ability to save anything, but I could do it.
The people tagging data are making much less.
Bro. $300k puts them into the top 2% of income in the US. It's expensive but not a struggle by any means.
But not for the Bay Area. As of November 2024, the median price of a home in the Bay Area is around $1,316,500.
On a 3:1 debt to income ratio, I'd put median income for the area for those who can afford to buy a home there is approx $350K (80% financing on median home price).
The idea that OpenAI had him killed is beyond ludicrous. Even if it were in any way realistic that OpenAI leadership somehow had access to hit men (!!), killing this guy would have been absolutely pointless as far as OpenAI's possible legal troubles go. The notion is not only outlandish, but also illogical.
He publicly stated OpenAI used copyright materials. What would killing him now possibly accomplish? Is there some legal statute I'm unaware of where a company can't be investigated if the original whistleblower is dead? It's not like he's the sole witness to a crime, and without his testimony the case disappears.
It’s improbable but not ludicrous.
The Steiners were harassed and threatened both online and physically in their home by deliveries of such things as a bloody pig mask, live cockroaches and spiders, a funeral wreath, and large orders of pizza.[5][1][6] Pornographic magazines with David Steiner’s name on them were sent to a neighbor’s house.[5][1][6]
Employees flew from California to Boston so they could vandalize the couple's Natick, Massachusetts home as well as stalk their personal vehicle.[1][6] Plans were even made to break into the couple's garage and place a GPS tracker on their car.[1]
https://en.m.wikipedia.org/wiki/EBay_stalking_scandal
This is just the incompetent tip of the iceberg.
Having access to hit men isn't actually that difficult, the hard part is telling which ones are undercover police.
Murderers don't act rationally. If someone at OpenAI did kill this guy, it wouldn't have been because they thought it was in their interest. It would have been because they had some kind of emotional problem or an extreme case of entitlement and thought they could get away with it. People in all walks of life will commit crimes, and guessing innocence on the basis of rational motives is about as reliable as guessing guilt.
> Even if it were in any way realistic that OpenAI leadership somehow had access to hit men
I don't think it's crazy to think people with that much money could find a way to have someone killed.
It's absolutely bonkers to think that the leadership of OpenAI thought to themselves: "This employee who spilled the beans on our use of copyrighted material, we better off him before he says more." That's just nuts. It's the weakest-sauce of possible things to blow a whistle on anyway, given how content owners are already pitchforks-out against AI companies since day one. He's a random employee who came out and said "That obvious thing you all think OpenAI is doing (slurping in copyrighted data)... yup they did that." And exactly how woult killing him would fix their problem anyway?
People here are indulging in fantasy.
It doesn’t have to be the CEO or the execs directly and there are many plausible stories that could have occurred.
What if it was a former coworker who is deeply in gambling debt and needs OpenAI to IPO in order for him to not lose everything in his life?
I think the people around the whistleblower would know better whether or not he was in the mental state to commit suicide. No note for his parents is pretty sus, especially given all the plans they made to visit.
I don’t think that they played a part in this, only that they have the means.
I can only imagine you being a "1st world" citizen for being this naive.
> "OpenAI somehow had access to hit men (!!)"
You think hitmen are only for movies?? Even much much smaller scale companies have people who sort out their problems. OpenAI is one of the biggest companies. Of course it has hitmen, connections with the police, FBI, CIA, NSA and whatever you can think of.
These companies (Boeing, OpenAI, Google, Microsoft) are sending a MESSAGE to the future whistleblowers. "You step on our tail, you're done" and they are making it as obvious as possible and you idiots still don't get the message?
You think these are some cute goofy nerds sitting and doing tech stuff and playing warcraft all day?
Absurd, what makes you think a criminal conspiracy of the scale you’re speculating about could possible stay secret, long term?
Testifying during a trial? I am not saying this happened, but it is not unheard of witnesses in high-profile trials to get murdered (and getting covered up) even if they have been public about it. Getting public about it means nothing in terms of testimonies during the trial, and if the person is dead there is no witness's testimony to be used.
I do not know about the specifics in "open"AI's court cases, but the fact if him being public in general does not mean much, as I understand it. Moreover, it can scare future whistleblowers. The motive is definitely there, now if this is what happened is a different story.
> "The idea that OpenAI had him killed is beyond ludicrous."
On the other hand, given that everyone either assumes the whistle-blower was murdered, or at least that the stress of the situation contributed to the suicide, why did OpenAI not do absolutely everything in their power to keep the person safe and mentally well?
Why should OpenAI have zero liability for the death, which anyone could have foreseen was a reasonably possible outcome?
It just seems like whistleblower protection laws could be much more effective if the company had genuine incentive in the person not dying.
Intimidation of future whistleblowers
I don't have any skin in the game as far as this particular case goes but if there was a list of CEOs who would order a hit, Sam Altman would 110% be near the top of that list. The guy is a straight-up sociopath.
> What would killing him now possibly accomplish?
Dissuading future potential whistleblowers of course.
He always makes very loud, bold statements to make himself look good and then quietly reverses them later on.
I don’t care about money. But he drives a $4 million car.
OpenAI will never be for profit. Actually we will create a for-profit wing.
I will never take equity. My investors want me to take an equity stake, not me!
OpenAI will not be used for weapons. OpenAI will be used for defensive weapons only.
Don’t believe anything he says at this point. He wants AGI and he wants to cut everyone else out and he wants to be in control of it and become the next richest and most powerful man in the world.
> The idea that OpenAI had him killed is beyond ludicrous
Regular people have hired contract killers to kill their partner, for completely pointless and stupid reasons.
What's ludicrous is your outright dismissal.
Suchir’s blog for reference - https://suchir.net/fair_use.html
as is their right. Companies should be held responsible for at least putting undue social, financial and legal pressure on people who are speak out for their beliefs regarding public matters.
Did OpenAI put social, financial, or legal pressure on him? Genuinely asking, I have no idea
> Did [anyone] put social, financial, or legal pressure on him? Genuinely asking, I have no idea
Great question. If the family has suspicion of such then perhaps a competent authority should perform an investigation.
Sure but the family is overly biased. They need something more tangible.
Does the family have a concrete basis for suspicion of such? The police don't have any particular power to investigate wild hypotheses. They can ask his friends and coworkers "hey did you kill him", and they'll respond with some combination of "no" and "talk to my lawyer" - then what?
The same gov that’s in bed with OpenAI???
In what way?
The head of the NSA left his job to join the board of directors of OpenAI.
Total coincidence, nothing to see here, peasants. If you think otherwise you’re a conspiracy theorist, a Putin apologist, a gaslighter, or whatever we’re calling you today.
[dupe] https://news.ycombinator.com/item?id=42524190
This comment is itself a dupe.
https://news.ycombinator.com/item?id=42535251
[flagged]
[dead]
What the hell is wrong with peoples brains? Everyone in these comments are like "Oh not OpenAI wouldn't do that", "maybe he had mental health issues" Shut up please. How naive are you people?
Magically every whistleblower is dying and they all have mental health issues? God must hate whistleblowers then.
Bots are real, man. HN's bots are top-notch. There is no other reason of these comments.
It's easier to believe companies are your friends than to consider that the elites would do anything to save their money.
Americans and Western Europeans are in general so damn naive when it comes up to that. Like sure, live in your fantasy where everyone abides the law and no bad actors exist :<
The amount of harassment and hate being propagated by the media certainly didn't help. I would also find The New York Times to be liable for all the AI doomerism they propagate.