Ronan Farrow here. Andrew Marantz and I spent 18 months on this investigation. Happy to answer questions about the reporting.
This is a fantastic piece, very timely, evidently well-researched, and also well-written. Judging by the little that I know, it's accurate. Thank you for doing the work and sharing it with the world.
OpenAI may be in a more tenuous competitive position than many people realize. Recent anecdotal evidence suggests the company has lost its lead in the AI race to Anthropic.[b]
Many people here, on HN, who develop software prefer Claude, because they think it's a better product.[c]
Is your understanding of OpenAI's current competitive position similar?
---
[a] You may want to provide proof online that you are who you say you are: https://en.wikipedia.org/wiki/On_the_Internet%2C_nobody_know...
[b] https://www.latimes.com/business/story/2026-04-01/openais-sh...
[c] For example, there are 2x more stories mentioning Claude than ChatGPT on HN over the past year. Compare https://hn.algolia.com/?dateRange=pastYear&page=0&prefix=tru... to https://hn.algolia.com/?dateRange=pastYear&page=0&prefix=tru...
https://xcancel.com/RonanFarrow/status/2041127882429206532#m
https://www.theguardian.com/technology/2026/feb/12/elon-musk...
xcancel is a valid workaround for X links on Hacker News and is sufficient for original attribution.
Unfortunately it probably doesn't even matter here on HN considering how brigaded down this story is predictably getting.
But yeah, it was a fantastic piece.
The piece captures some of the anxieties within OpenAI right now about their competitive position. This obviously ebbs and flows but of late there has been much focus on Anthropic's relative position. We of course mention the allegations of "circular deals" and concerns about partners taking on debt.
I was asking more about your informed view on how OpenAI's technology, products, and roadmap are perceived, particularly by customers and partners, in comparison to those of competitors.
If you have an opinion about that, everyone here would love to hear about it.
As it turns out, and what I’m kind of going with for this LLM shit, is that it’ll play out exactly how you think it will. The companies are all too big to fail, with billionaire backers who would rather commit fraud than lose money.
If you're picking on my vocabulary, that's fair. Fraud wasn't the point, I think you're smart enough to realize that.
Trading shares for GPUs is not corrupt either.
He's had so many conversations that he likely has a sense of how perceptions of the company and its offerings have changed.
I'm curious.
I.e. what I used to use Google for and when I don't want an AI to overly summarize / editorialize result data.
The problem is sometimes on paper everything people like Sam Altman do is legal, despite it harming so many. We've literally had a major RAM producer pull off the consumer RAM market. I feel like Sam Altman should be investigated and heavily scrutinized. He kind of is the biggest bubble in the AI bubble, we're letting him fester too far into it too, and these circular deals have seemingly somewhat stopped for now, but it might only get worse.
I guess my question was more, if the article author was the judge of fate or morality, what should happen?
As to AI and Sam, I think it’s too early to tell what effects will be. So we should adopt non judgement, build good ourselves and see what unfolds.
No comment on the CEO: I just find the product superior in everything but UI/UX and conversation. It's better at quality code.
Both codex and Claude code fail when it comes to extremely sophisticated programming for distributed systems
Also. RLHF mean that models spit out according to certain human preference, so it depends what set of humans and in what mood they've been when providing the feedback.
In general, Claude was impressed by what Codex produced and noted the parts where it (i.e. Claude) had missed something vs. Codex "thinking of it".
From a "daily driver" perspective I still use Claude all the time as it has plan mode, which means I can guarantee that it won't break out and just do stuff without me wanting it to. With Codex I have to always specify "Don't implement/change, just tell me" and even then it sometimes "breaks out" and just does stuff. Not usually when I start out and just ask it to plan. But after we've started implementation and I review, a simple question of "Why did you do X?" will turn into a huge refactoring instead of just answering my question.
To be fair, that's what most devs do too (at least at first), when you ask them "Why did you do X" questions. They just assume that you are trying to formulate a "Do Y instead of X" as a question, when really you just don't understand their reasoning but there really might be a good reason for doing X. But I guess LLMs aren't sure of themselves, so any questioning of their reasoning obliterates their ego and just turns them into submissive code monkeys (or rather: exposes them as such) vs. being software engineers that do things for actual reasons (whether you agree with them or not).
Gemini seems to be the worst of the three, and some open-weight models are not too bad (like Kimi k2.5). Cursor is still pretty good, and copilot just really really sucks.
I enjoy using CC more and use it for non coding tasks primarily, but for anything complex (honestly most of what I do is not that complex), I feel like I am trading future toil for a dopamine hit.
LLMs aren't able to achieve 100% correctness of every line of code. But luckily, 100% correctness is not required for debugging. So its better at that sort of thing. Its also (comparatively) good at reading lots and lots of code. Better than I am - I get bogged down in details and I exhaust quickly.
An example of broad work is something like: "Compile this C# code to webassembly, then run it from this go program. Write a set of benchmarks of the result, and compare it to the C# code running natively, and this python implementation. Make a chart of the data add it to this latex code." Each of the steps is simple if you have expertise in the languages and tools. But a lot of work otherwise. But for me to do that, I'd need to figure out C# webassembly compilation and go wasm libraries. I'd need to find a good charting library. And so on.
I think its decent at debugging because debugging requires reading a lot of code. And there's lots of weird tools and approaches you can use to debug something. And its not mission critical that every approach works. Debugging plays to the strengths of LLMs.
For the few times I've used both models side by side on more typical tasks (not so much web stuff, which I don't do much of, but more conventional Python scripts, CLI utilities in C, some OpenGL), they seem much more evenly matched. I haven't found a case where Claude would be markedly superior since Codex 5.2 came out, but I'm sure there are plenty. In my view, benchmarks are completely irrelevant at this point, just use models side by side on representative bits of your real work and stick with what works best for you. My software engineer friends often react with disbelief when I say I much prefer Codex, but in my experience it is not a close comparison.
Is there one that you prefer for, i dunno, physics?
That said, I am puzzled at the algorithms that Claude & GPT "get" and ones that they do not.
(former physicist here. would love to know the kind of things you're working on. email on my profile)
I've been working on a wide range of relatively projects and I find that the latest GPT-5.2+ models seem to be generally better coders than Opus 4.6, however the latter tends to be better at big picture thinking, structuring, and communicating so I tend to iterate through Opus 4.6 max -> GPT-5.2 xhigh -> GPT-5.3-Codex xhigh -> GPT-5.4 xhigh. I've found GPT-5.3-Codex is the most detail oriented, but not necessarily the best coder. One interesting thing is for my high-stakes project, I have one coder lane but use all the models do independent review and they tend to catch different subsets of implementation bugs. I also notice huge behavioral changes based on changing AGENTS.md.
In terms of the apps, while Claude Code was ahead for a long while, I'd say Codex has largely caught up in terms of ergonomics, and in some things, like the way it let's you inline or append steering, I like it better now (or where it's far, far, ahead - the compaction is night and day better in Codex).
(These observations are based on about 10-20B/mo combined cached tokens, human-in-the-loop, so heavy usage and most code I no longer eyeball, but not dark factory/slop cannon levels. I haven't found (or built) a multi-agent control plane I really like yet.)
I do regular evaluation of both codex and Claude (though not to statistical significance) and I’m of the opinion there is more in group variance on outcome performance than between them.
Last one is from yesterday: https://news.ycombinator.com/item?id=47660925
Codex has been consistently better on almost every level.
* (an open source framework for 2D games in Godot 4.6 GDScript, mostly using AI to review existing code)
There are two types of vaccine be coders. Those who review the code generated and those who don’t.
Either because they don’t understand code at all, or because they don’t have time and don’t care.
Code quality is only one factor. Naive vibe coders, who don’t code otherwise, rate performance based on output alone.
(i found your comment surprising based on my daily hn reading recollection - i mostly read top N daily and feel i only occassionally see codex stories).
Did you do any extra investigations into Annie’s allegations? It feels to me like the unstated conclusion is recovered memory can’t be trusted, which is a popular understanding but a very wrong one put out by the now defunct and discredited False Memory Syndrome Foundation. It was founded by the parents of the psychologist who coined DARVO, directly in reaction to her accusing them of abuse.
Dissociation is real (I have a dissociative disorder, and abuse I “recovered” but did not remember for much of my adolescence and early adulthood has been corroborated by third parties) and many CSA survivors have severe memory problems that often don’t come to a head until adulthood. I know you didn’t dismiss her claim, but the way the public tends to think about recovered memories is shaped primarily by that awful organization.
As noted in the piece, we spent months talking to Altman's partners and what we found and didn't is as described.
There’s a lot more who remember and may not have corroboration more than with themselves and among their close friends or healthcare provider. Part of CSA is usually there is very little a kid can do about evidence, as the power discrepancy is far too much. Often with rich abusers, the exact same process occurs. Perps pick victims who are vulnerable or controllable, and constantly seek power and domination. Nothing to do with the boardroooms or batch of ceo billionaires running the economy right now certainly.
That said, "recovering" memories as a therapy does not pass any sort of sniff test and it doesn't take a concerted effort to discredit the concept. Human memory is very malleable. Patients with mental health issues (which could predate abuse, or could be caused by abuse) are often in search of answers and that makes them very vulnerable.
Could a memory be buried deep in our subconscious, forgotten, only to return to the surface later? Sure, we all forget things and then remember them when triggered by something, whether that's a smell or sound or something else entirely. But can we engineer that process, with any degree of reliability? How can we even begin to reliably reverse engineer the triggers?
I think it is also important to keep in mind that Annie is rich, and the health care available to rich people can be very predatory. There are endless examples of nonsense therapies for all types of health, from ear seeds to treatments for "chronic Lyme".
Memories that return organically due to a trigger are a world apart from "recovered" memories, we shouldn't conflate them. If Annie's memories were triggered in adulthood, sure, that's really no different than remembering something... but "recovered"? That is something else entirely.
Correct me where I'm wrong, I'd like to learn your perspective, maybe there's a missing piece.
Recovered memory therapy was a discredited hypnotherapy that leaned heavily on suggestion or was associated often with fairly coercive interrogations during the 80s CSA panic - https://en.wikipedia.org/wiki/Day-care_sex-abuse_hysteria
> Memories that return organically due to a trigger are a world apart from "recovered" memories, we shouldn't conflate them.
Agree, though I think the mechanism can be a bit more towards the idea of a “recovery” of traumatic memory, even if the term as understood carries false connotations.
The concept you’re missing is dissociation, and dissociative disorders. In the 40s it was called just “hysteria”, and for many cases up to the late 90s an extreme form was called multiple personality disorder, now DID (dissociative identity disorder). https://en.wikipedia.org/wiki/Dissociative_disorder
Not everyone who goes through traumatic events will respond to it via dissociation of identity, and indeed not all people are equally capable of developing a dissociative disorder, 2 people may go through very similar events (say survive a war as siblings or even twins) and one might dissociate the traumatic experience and one might not. Dissociation doesn’t work quite like you might imagine from a term like “multiple personalities” - that happens in some extreme cases, but think of identity dissociation as an adaptive response to events or situations that are paradoxical (esp to a child’s mind), extreme or traumatic, and can’t be escaped or use of other mechanisms cant be called upon.
Dissociation is on a sort of spectrum, where at one side you have common experiences like zoning out when on a common commute, and on another you have separated self-parts/alter egos to handle wildly different situations.
It’s a mechanism I frankly wasn’t aware of and I’m not sure that I would be able to fully beleive or empathize with, but for my getting a diagnosis of a dissociative disorder changed my life, and made a thousand things about me that I could never figure out make sense. The “model” as it put it at the time responded to experiment, and by recognizing that I was dealing with pretty constant, heavy dissociation and different self states with memory deficiencies helped me figure out how to work through a ton of really intractable problems for me. I’m finally after decades of ineffective therapy able to really understand how I work.
Idk how to talk about it without sounding like I’m trying to sell the idea. But yeah it was a mind blowing thing to me. Over the last 20 years especially a ton of truly respectable research has been done and the increase in efficacy of treatments on dissociation, and trauma generally is one of the unsung advancements for humanity in the last decade. I think the number is that around 3-6% of people meet the clinical criteria for a dissociative disorder - OSDD, DID, DPDR, or dissociative amnesia. 5x more people than have schizophrenia, 5x more than have red hair.
My favorite public clinical resource I point to people is the CTAD Clinic YouTube - https://youtube.com/@thectadclinic?si=5AyR5H8K8Cf2sn3C
Pretty easy to understand explainers from a clinician in the UK.
For a more clinical and study approach this one is the currently best put together research IMO: https://www.taylorfrancis.com/books/edit/10.4324/97810030573...
The TLDR is dissociation is an important mechanism that most people don’t know about but has had a wave of research and study and is much more common than one might expect. The sad part is how often dissociative disorders correlate w abuse.
I’m reading more now and I think the missing piece for me is the distinction between “repressed” memories and “recovered” memories.
I understood repressed memories to be an accepted idea, distinct from “recovered” memories. I am reading that the people mentioned in your original comment rejected the idea of repressed memory altogether, and believed that everything traumatic must be remembered.
So, to me, reading that someone “recovered” memory reads like they went through a specific type of therapy intended to “find” these repressed memories. Whereas to you, “recovered” memories could be repressed memories that came back to the surface organically — whether at random, triggered or through a therapy intended to deal with disassociating. Is that right?
> It feels to me like the unstated conclusion is recovered memory can’t be trusted, which is a popular understanding but a very wrong one put out by the now defunct and discredited False Memory Syndrome Foundation.
with
> Recovered memory therapy was a discredited hypnotherapy
I read your first post as standing up for recovered memory therapy and I can't find how the discussion of dissociation makes a difference. Does Fontain have it right that by "recovered memory" you mean "things people happened to remember on their own"?
My question is, how do you know when an enormous project like this, conducted over an 18-month time span is "done"? I assume you get a lot of leeway from editors and publishers on this matter. How do you make the decision to finally pull the trigger on publishing?
You try to reach a critical mass of detailed, rounded understanding of a central question, integrating the most meaningful perspectives, interrogating the weak points and blind spots, and backing up the assertions with documentary evidence or strong sourcing. Eventually, you reach a point where enough sources and materials are reliably triangulating toward the same truths.
As you guessed, there's external pressures that figure in this analysis—whether competitors are closing in on the same leads; what's happening in the broader news cycle that might make a story feel more or less relevant. As you also guessed, I am more fortunate than most writers in the degree to which I get to hold off until something feels fully baked. Mostly, writers simply have to hit a deadline, and resources run out before ambition does. I have deadlines and constraints too, but I get a lot of say in how I organize all of the above.
Then there's the actual process of creating the story. Writing a densely evidence-based investigative piece is labor-intensive—in this case, weeks of initial drafting, and then much iteration. The fact-checking process at the New Yorker is exhaustive, and can span weeks. Every sentence, assertion, and piece of underlying sourcing get scrubbed by multiple independent pairs of eyes. This story had four fact-checkers working on it for the better part of a two week period, pulling very long hours. This is all brought together in a closing meeting where each sentence is revised and polished in a group.
This is all done as additional information comes in—in fact, with these large-scale bodies of reporting, there is very often a snowball effect, where a lot comes in at the end.
I would really suggest subscribing to and finding ways to amplify independent outlets and journalists, and encouraging others to do so.
In developer communities often you can support individual developers or groups through a monthly subscription / donation on their github page or similar.
The Sam Altman piece can be read here: https://apple.news/APTX4OkywRWeJXIL7b8a7zQ
As for your #2, that seems reminiscent of "why are we going to space when there are so many problems here on Earth."
(2) I'm sorry the post was downranked off the frontpage for a while this afternoon. A software penalty kicks in when the discussion seems overheated ("flamewar detector") but I turned this off as soon as I became aware of it. We make a point of moderating HN less when a story is YC-related (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...) but as this goes against standard internet axioms, people often assume the opposite.
(* And yes, any reader who wants this is welcome to email hn@ycombinator.com to ask - I haven't turned it on for everyone because I'm worried it would slow the site down. Also, it's a bit buggy and not only have I not had time to fix it, I've forgotten what the bugs are.)
Can't wait until this is released!
https://theonion.com/anyone-else-have-those-weird-dreams-whe...
https://theonion.com/the-onions-exclusive-interview-with-sam...
Includes gems such as:
Q: What informs your personal sense of morality?
A: Previous things I’ve gotten away with.
Q: Why did you decide to devote your life to AI?
A: I just saw so much suffering in the world that needed to be automated.
All evidence today suggests Anthropic is passing OpenAI in relative and absolute growth. So where's the critical reporting? The DOD coverage was framed around the Pentagon's decisions, not Anthropic's. And nobody seems interested in examining whether the company that branded itself as the ethical AI lab actually is one. That seems like a story worth writing.
Now, they may have heard the word "Anthropic" due to recent media coverage. But they don't know what it is and don't remember what it makes. The fact that all businesses use "Anthropic" is about as relevant to them as knowing the overseas shipping company for all the shit they buy off Amazon.
So articles about OAI will always produce more revenue for the media, because it's related to what normies actually use day to day.
That really rubbed a lot of people the wrong way, as ultimately one might have a favorite tool, then suddenly they are forced to use another tool.
You have a point in that Anthropic deserves some coverage too and that there are interesting perspectives that we've not heard of on that front either.
But just because that's true doesn't mean this article isn't very much relevant and needed.
Because it is.
FWIW I have two(!!) close friends working for Anthropic, one for nearly two years and one for about 4 months.
Both of them tell me that this is not just marketing, that the company actually is ethical and safety conscious everywhere, and that this was the most surprising part about joining Anthropic for them. They insist the culture is actually genuine which is practically unicorn rarity in corporate America.
We have worked for FAANG so I know where they're coming from; this got me to drop my cynicism for once and I plan on interviewing with them soon. Hopefully I can answer this question for myself.
From the outside, I find Anthropic's hyperbolic marketing to be an indication that they are basically the same as every other bay area tech startup - more or less nice folks who are primarily concerned with money and status. That's not a condemnation, but I reject all the "do no evil" fanfare as conveniently self serving.
This isn't remotely true in my experience. The senior folks I know at Meta, for example, pretty much concede they're ersatz drug dealers.
Google was "do no evil" until they had to choose between that and making the money. The culture has to be not only professed but tested.
And there's something true there; few companies are Snidely Whiplash evil (maybe the lawnmower but even that is just what it is) - and having large amounts of cash affords you options in many areas.
Certainly most of us know we are just in it for the money, and the soul-grinding profit machine will continue to grind souls for profit regardless of what we want.
So that's why it is surprising to me when my (fairly senior) grizzled ex-FAANG friends, that share the same view, start waxing poetic about Anthropic being different and genuine. I think "maybe it is" and decide to interview. IDK, I guess some part of me wants to believe that nice things can exist.
If the leadership doesn’t bend it might get replaced. It’s annoying. I think Claude is atm the best AI assistant, by far.
It only showed they were marginally more ethical than OpenAI and XAI which isn't saying much.
It's unfair to sweep provision of methods to the military under a "respect the service" catch-all justification.
Two things can simultaneously be true: (1) individuals serving in the military are making sacrifices (in terms of pay, family life, personal safety) that deserve respect and (2) the military as a political institution will amorally deploy whatever capabilities it has access to, to achieve political aims.
There's a reason the US stopped offensive chemical, biological warfare, and tactical nuclear device research and production -- effective capabilities will be used if they exist.
Just a couple of thoughts since it seems like the next issues in this space are rapidly arriving or already here.
It was also the product of perceived overmatch on both sides -- the Soviets believed they had superior mass of armored formations (and they did), while the US and allies believed they had technological supremacy (and they did). Ergo, neither needed tactical nukes.
It didn't hurt that it helped both in the eyes of the then vehemently anti-nuclear European movements.
Offensive bio and chemical weapon limitation is a more nuanced decision.
In both cases, their primary use was either local mass lethality or terrain denial, neither of which were important in the then-gelling American doctrines of maneuver.
The sole use case they seemed viable for was industry denial (e.g. contaminate a high capital cost industrial center), a task at which strategic sized nuclear weapons were equally adept (and more easily stored). So, if you had to have strategic nuclear weapons for deterrence, and they were capable of the same task, why have fiddly bio and chemical weapons?
But in both cases there was also a constant radiant pressure of scientists and the public campaigning against them, and being unwilling to work on or tolerate them.
Absent that, who knows how history would have turned out? Normalization is a powerful opinion shifter.
So much so that I worry they won't be Machiavellian enough to survive. Hope I am wrong.
I wonder what Anthropic tries to achieve by spreading such blatant lies with their bot accounts. I'm definitely not buying anything from a company so morally corrupt to smear the competition while claiming to be somehow "ethical". And I'm not talking just about this thread, it's a recurring pattern on Reddit.
[1] "Unless Its Governance Changes, Anthropic Is Untrustworthy" https://anthropic.ml/
Anthropic is emphatically not safe. None of the AI labs with customers (i.e., excluding a few small nonprofits whose revenue comes from donations) are anything like safe -- because of extinction risk. The famous positive regard that Anthropic employees have for their organization's mission means almost nothing because there have been hundreds of quite destructive cults and political parties whose members believed that theirs is the most ethical and benign organization ever.
The best thing you can say about Anthropic is that if you have to support some AI lab by becoming a customer, investor or employee, it is slightly less dangerous for the world to support Anthropic than OpenAI although IMHO (and I admit I am in a minority on this among extinction-risk activists) it is slightly less dangerous to support Google Deep Mind or Mistral than Anthropic.
All four organizations I mentioned should be shut down tomorrow with their assets returned to shareholders.
The current crop of services provided by the leading AI labs are IMHO positive on net in their effect of people and society, but the leading AI labs are spending a large fraction of the 100s of billions of dollars they've received from investors on creating more powerful models, and they might succeed in their goal of creating models that are much more powerful than the ones they have now, which is when most of the danger would manifest.
The leaders of all of the leading AI labs have the ambition of completely transforming society and the world through AI.
That would be irrational.
We should give air time to other problems?
I think everyone agrees with that.
You have managed to distill a surprisingly pure vintage of false dichotomy, from a near Platonic varietal of whataboutism.
Not making any value judgements, but I can see how one might value their interpretability research higher than what the ceo says in a time where the corrupt, criminal executive branch is muscling in to everything from what's written on currency, to journalistic sources. I generally blame fascists before i blame those unable or unwilling to resist them. though obviously, ideally, we'd all lock arms and, together through friendship, crush authoritarians and fascists.
“I had the burden of impacting public affairs through my wildly succesful corporation”—poor them.
For me, a big worry about AI is in its potential to further ease distorting or fabricating truth, while simultaneously reducing people's "load-bearing" intellectual skills in assessing what is true or trustworthy or good. You must be in the middle of this storm, given your profession and the investigations like this that you pursue.
Do you see a path through this?
As a reader, am I supposed to infer anything about evidentiary weight from these stylistic choices? When a single anonymous source's testimony is presented in a "declarative" narrative style like here (with the attribution in a less prominent position), should we read that as reflecting high confidence on your end (perhaps from additional corroboration not fully spelled out)? And does the fact that Altman’s non-recollection appears in parentheses carry any epistemic signal (e.g. that you assign it less evidentiary weight)? Or is that mostly a matter of (say) prose rhythm?
I’m sure you don’t know half of the totally fucked up things Sam did to get “revenge” for the slight of a leaking pool.
> [Graham's] judgment was based not on Altman’s track record, which was modest, but on his will to prevail, which Graham considered almost ungovernable.
One thing I don't understand is why Paul Graham offered YC to Altman if he knew how slippery he was..
Altman describes his shifting views as genuine good faith evolution of thinking. Do you believe he has a clear North Star behind all this that’s not centered on himself?
My own impression after many hours of conversation is that he is identifying something of a true north star when he frames this around "winning." There are people in the story who talk about him emphasizing a desire for power (as opposed to, say, wealth). I think he probably also believes, to some extent, the story he tells that equates winning, and his gaining power, with a superabundant utopian future for all.
However, I think critics correctly highlight a tension between his statements about centering humanity writ large and his tilt into relentless accelerationism.
Also a good expose on accelerationists and e/accs and who among the elites fall in this group is direly needed as well
Is this something you've thought about?
Long time HN lurker, made an account just to say that :)
It's not your responsibility to fund for every single one, just find the one you like the most and subscribe to that one.
I have not read the article yet, because I get the physical magazine and look forward to reading it analog. I therefore only have an inconsequential question.
I love the New Yorker’s house style and editorial “voice,” and I have always been curious about the editing process. I enjoyed the recent exhibit at the NYPL, which had some marked up drafts with editor feedback and author comments.
Did you find that your editors made significant changes to the voice of the piece, and/or do you find any aspects of their editing process particularly notable or unusual?
Can’t wait to read this one, and hope the HN crowd treats you well.
Thank you for fielding questions. And please don't stop, your work is great.
“Tonight isn’t just about the people in front of the camera. In this room are some of the most important TV and film executives in the world. People from every background. But they all have one thing in common: They’re all terrified of Ronan Farrow.”
Clearly he's straight up evil; between tanking the global economy, constantly lying, and raping his 3 year old sister, it feels really disingenuous to me to frame this as an open question.
I’m far more concerned with the 25 million dollar personal bribe OpenAI president Greg Brockman gave Donald Trump for his reelection -
the fact that a tech company can influence the outcome of an election directly is evil
Far more evil than Altmans shenanigans
People want to focus on scape goats rather than systemic problems
My prima facie view on Altman has been that he presents as sincere. In interviews I have never seen him make a statement that I considered to be a deliberate untruth. I also recognise that people make claims about him go in all directions, and that I am not in a position to evaluate most of those claims. About the only truly agreed upon aspect has been how persuasive he is.
I can definitely see a possibility of people feeling like they have been lied to if they experienced a degree of persuasion that they are unaccustomed to. If you agree to something that you feel like you didn't really feel like you would have, I can see people concluding that they have been lied to rather than accept that they had been intellectually beaten.
In all such cases where an issue is contentious, you should ask yourself, what information would significantly change your views. If nothing could change your view, then it's a matter beyond reason.
I think you will agree that there is no smoking gun in this article, and it is just an outlay of the allegations. Evaluating allegations becomes tricky because I think it becomes a character judgement of those making the claims.
I have not heard a single person in all of this criticise Ilya Sutskever's character. If he were to make a statement to say that this article is an accurate representation of what he has experienced, it would go a long way.
I think Paul Graham should make a statement, The things he has publicly claimed are at odds with what the article says he has privately claimed. I have no opinion if one or the other is true or if they can be reconciled but there seem to be contradictions that need to be addressed.
While I do not have sources to hand (so I will not assert this as true but just claim it is my memory) I recall Sam Altman himself saying that he himself did not think he should have control over our future, and the board was supposed to protect against that, but since the 'blip' it was evident that another mechanism is required. I also recall hearing an interview where Helen Toner suggested that they effectively ambushed Altman because if he had time to respond to allegations he could have provided a reasonable explanation. It did not reflect well on her.
I am a little put off by some of the language used in the article. Things like "Altman conveyed to Mira Murati" followed by "Altman does not recall the exchange" Why use a term such as 'conveyed' which might imply no exchange to recall? If a third party explained what they thought Altman thought. Mira Murati could reasonbly feel like the information has been conveyed while at the same time Altman has no experience of it to recall. Nevertheless it results in an impression of Altman being evasive. If the text contained "Altman told Mira Murati" then no such ambiguity would exist.
"Later, the board was alarmed to learn that its C.E.O. had essentially appointed his own shadow board" Is this still talking about Brockman and Sutskever? I just can't see this as anything other than a claim he took advice from people he trusted. I assume those board members who were alarmed were not the ones he was trusting, because presumably the others didn't need to find out. The people he disagreed with still had votes so any claim of a 'shadow board' with power is nonsense, and if it is a condemnable offence, is the same not true of the alignment of board members who removed him.
Josh Kushner apparently made a veiled threat to Muratti, the claim "Altman claims he was unaware of the call" casts him as evasive by stacking denial upon denial, but without any other indication that was undisclosed in the article, it would have been more surprising if he did know of the call. I also didn't know of the call because I am not those two people.
The claim of sexual abuse says via Karen Hao "Annie suggested that memories of abuse were recovered during flashbacks in adulthood." To leave it at that without some discussion about the scientific opinion on previously unremembered events being recalled during a flashback seems to be journalistically irresponsible.
> I can see people concluding that they have been lied to rather than accept that they had been intellectually beaten.
There are two angles to this: from an individual perspective and from a collective one.
One's interaction with such a manipulator isn't a single shot. There is not a single event that they are “beaten”. First, one gets persuaded --- you might argue that there's nothing wrong with a skillful persuasion. At some point they realize that the reality is not in line with their expectations. They bring the point up to the manipulator and ask for a change, this time in more concrete terms. The manipulator agrees with the change, negotiates compromises, and the relationship continues. After some time the manipulated party realizes that things are not going in the direction they desire. This time they ask for more concrete terms, without accepting any compromises. The manipulator accepts, yet continues to act against the terms. The manipulated party is now angry and directly confronts the manipulator. The manipulator apologizes and tells that none of it was intentional, and asks for another chance. However, at that point, the manipulator has run out of “politically correct” “persuasion tactics”, and tells blatant lies to make the other party behave.
From a collective perspective, even those “politically correct” “persuasion tactics” are discovered to be lies, because what the manipulator told different parties are in direct opposition to each other, i.e., they cannot all be truths.
> Helen Toner suggested that they effectively ambushed Altman because if he had time to respond to allegations he could have provided a reasonable explanation. It did not reflect well on her.
I understand how her behavior may raise a flag for the unsuspecting, but it was exactly the right one. Manipulators prey on the benefit of the doubt. If Toner were to bring Altman's behavior into attention of others, no doubt that Altman would manipulate them successfully.
It's unfortunate that many people are unaware of these tactics and assume the best of intentions, when such assumptions fuel the manipulation that they would better avoid.
Here is the etymological definition of the word:
persuasion(n.) late 14c., persuasioun, "action of inducing (someone) to believe (something) by appeals to reason (not by authority, force, or fear); an argument to persuade, inducement," from Old French persuasion (14c.) and directly from Latin persuasionem (nominative persuasio) "a convincing, persuading," noun of action from past-participle stem of persuadere "persuade, convince," from per "thoroughly, strongly" (see per) + suadere "to urge, persuade," from PIE root *swād- "sweet, pleasant" (see sweet (adj.)).
Meaning "state of being convinced" is from 1530s; that of "religious belief, creed" is from 1620s. Colloquial or humorous sense of "kind, sort, nationality" is by 1864.
IMHO if you aim to convince people of something you are on the side of trying to control people's freedom to chose. That in itself is a form of being unethical to the idea of truth.
If you can't let people come to their own conclusions, you got problems and you shouldn't be in a position of power.
In my experience the people who spend the most time convincing are people with narcissistic personality disorders. I stay far away from those people because I know they dont really value truth and justice like I do.
The industry seems to attract people who can only be described as smooth opportunists, always a shy step away from becoming confidence artists. People with big dreams of material success, but with next to no ability or talent, and with a tragic lack of self knowledge (and often a lack of domain knowledge). Pure entitlement and greed, and a desire to use other people as a bridge to the stars[0].
I will say this, however: they do have a keen sense of what the incentives are. They will keep doing this, for as long as society keeps rewarding them, and refuses to punish them. And unhooking those incentives is not difficult: do not let them externalize blame, do not let small dishonesties pass, do not let them internalize praise that belongs to someone else, and, most importantly, do not look the other way, when they decide to cannibalize the career of someone else, in order to nourish their own.
Silicon Valley, since at least the Web-2.0 days, has been about nerds making frat-bros rich, in exchange for a livable wage (salaries tend to be only slightly in excess of cost-of-living, unless you are willing to live either far away from your workplace, or in a small moldy-smelling box of a studio). This is a bad deal. Silicon Valley idolizes Steve Jobs (even when he was alive), but gives little thought to Steve Wozniak (upon whose work, Apple and the PC industry were built). When I was in college, both Jobs and Dennis Ritchie died within a short time of each other. Silicon Valley mourned Jobs, but only a few of us nerds mourned Ritchie.
Silicon Valley chose to name the most innovative car company in the country "Tesla", but it attracts and cultivates legions of future Edisons and Morgans[1]. And that is perhaps the perfect allegory for Silicon Valley: a car company named after someone like Tesla, but owned and operated by someone like Elon Musk[2].
[0]: Maybe this is an ancient human affliction. Did the pharaohs not do the same in their day?
[1]: Funny parallel, that since 2009, SV has been trying to rent out compute and storage, instead of just selling it outright.
[2]: Another pretender, seduced more by the warm glow of gold, than the cool blue crackling lightning of a Tesla Coil.
Incredible.
Quite simple: show me any single action took by Sam Altman which can not be construed as an attempt to get him more power/money/influence. You can't find it.
The difference between what he claims to believe and what he actually does is a textbook example of sociopathy.
"If you give me six lines written by the hand of the most honest of men, I will find something in them which will hang him"
To play your game, he got married, had a child, and joined an AI research organisation at a time when everybody thought the big advances were much further away than they turned out to be.
You could still construe those actions as evil if you choose to see them as evil.
I'm not going to claim that Sam Altman is not a sociopath, I lack the information and knowledge of psychology to make that determination. On the other hand I have not detected those attributes in anyone who has claimed he is a sociopath.
It seems odd that people seem to take offense at the notion that arbitrary people do not reach a conclusion that requires specialised expert knowledge and a decent amount of irrefutable evidence.
Try the other way around, via negativa. We definitely can find plenty of examples of people stepping out of positions of power, deciding not to do something because of moral conflict, etc. Is there any case of such action from Sam?
Fuck, anyone with any semblance of moral fortitude would refuse to take money from the Saudis. But he had no problem to do it.
> joined an AI research organisation at a time when everybody thought the big advances were much further away than they turned out to be.
No, this is selection bias. What he did was to put himself in a position where he could have his fingers on any and every possible pie, and then when of these things turned out to be something believed to be valuable by people with money, then he manouvered himself to be in the driver seat.
Thank you I was just going to point this out.
What you are thinking of are people who do bad things. Most of those people are not sociopaths. Often they are hurting in some way, sometimes they are just living the only life they know. Most of them feel they are doing the best they can. It is extremely comforting to pathologize these people because then it's not something we could have prevented by providing a better society. It rules out the hard options of empathising with them, or reasoning with them to find common ground.
The term othering has come into use in recent years. The concept has existed through the ages, but that's the latest label for it.
This is what it looks like.
It clarifies he did not fire Sam
I overall agree with your takeaway, but this is not a criticism of the article itself.
That is how pathological liars present.
I am also aware that sincere people present that way.
I don't believe there is any rational way to consider the appearance of innocence as evidence of guilt.
What other point could you have been making? You made no reference to any other evidence.
>as much as I admire the creativity in your interpretation, Mr. Self-Described Altman Apologist).
I am unsure if this is deliberate irony, or poor comprehension.
It may be more of a mental block than anything else.
And if this tendency to misunderstand/be misunderstood always results it Altman gaining more power, even if we give him the reason of the doubt and say that doesn't do it on purpose, it's still a big problem, given the responsibility he has.
The article also mentions many moments where apparently Altman straight out lied, as opposed to being "very persuasive, if you believe those sources then I don't think it's also possible to think he's sincere. I cannot open the article again to get the exact quotes, but the few I remember were: - one time he was claiming he didn't send a message, while people were literally showing him the message he sent, with the confirmation of another OpenAI employee - another time when he accused people of organising a coup, and that someone from the board informed him, and after the person from the board was called in the meeting Altman claimed he never said those words and never accused anyone
These cases can't be put to persuasion, that Altman changed their view, or that someone misremembered, they either happened or they didn't
Yes, but that doesn't work if you look for patterns selectively. There are large amount of people who will tell you vastly different experiences that they had with Altman. If you pick the right grouping, within it, you can find universal praise or condemnation. The article itself acknowledges that.
>The article also mentions many moments where apparently Altman straight out lied.
Does it? It has people saying he lied, and a few things he disputes that he said. If the lies were clearly apparent, I think his position would not be tenable. Which points in the article do they show statements that it clear that he has said them, that they were false, and that he knew they were false when he said them?
The points you list are not clearly apparent lies. At most they are allegations of lies. They might just be different interpretations of the same events. I have seen instances in my own life where someone has said "You said X" the other person says "No I didn't", The first then pulls up the minutes, and says "See you said X", the other responds with "That's not what that says". You see rage bait posts about terms and conditions that take that form all the time. Someone misreads a legal term as meaning something different to what it means in a legal sense and then refuses to acknowledge the commonly accepted definition.
Please respond to this, because I really am interested in the answer, but I did read the article and I didn't see what you appear to have seen.
I have made no claim to the merits of Sam Altman, I just don't like the idea of condemning someone on hearsay and insinuations. There are videos on YouTube claiming he's had people killed. At some point you have to point at something that everyone can agree on is an actual thing that happened and that it actually matters. At most what I have seen is people being able to provide one of those two points on any particular allegation.
I don't feel this should be that contentious. If it were clear there would be demands from all around saying "You did this bad thing, you must resign". Do you think that everyone dealing with OpenAI acknowledges some dark truth and is complicit?
The pieces in the article I was referring to are:
> Amodei’s notes describe escalating tense encounters, including one, months later, in which Altman summoned him and his sister, Daniela, who worked in safety and policy at the company, to tell them that he had it on “good authority” from a senior executive that they had been plotting a coup. Daniela, the notes continue, “lost it,” and brought in that executive, who denied having said anything. As one person briefed on the exchange recalled, Altman then denied having made the claim. “I didn’t even say that,” he said. “You just said that,” Daniela replied. (Altman said that this was not quite his recollection, and that he had accused the Amodeis only of “political behavior.”)
> Amodei discovered that a provision granting Microsoft the power to block OpenAI from any mergers had been added. “Eighty per cent of the charter was just betrayed,” Amodei recalled. He confronted Altman, who denied that the provision existed. Amodei read it aloud, pointing to the text, and ultimately forced another colleague to confirm its existence to Altman directly. (Altman doesn’t remember this.)
I agree it's very easy for 2 different people to understand or to remember something differently, and that meeting minutes are not always a reliable source, but for me in the 2 scenarios above is almost impossible for 2 people in good faith to disagree:
In the first case, if you say something, and a big deal is made of it, and 5 minutes later the other person claims that you said some specific words and you deny it, then someone is lying, either you or the other person.
In the second case, if there is something written in a contract, and someone presents that contract to you, reads it out loud, and asks a collegue to confirm, either that person made up the provision, or you are lying, there is little room for misunderstanding.
Given there are no proofs, I can't say he's 100% culprit, and I appreciate your rigor on this because we don't want to result judging everyone by a sort of "trial by public opinion".
However, outside of trials, the judjment can be more nuanced than a boolean "culprit/innocent", and to me the reasons below(*) are enough to distrust Altman and to prefer he wasn't the person at the head of a revolutionary technology that could have huge negative consequences on the society, or on human kind as a whole.
(*) the reasons being:
- amount of people interviewed and their very similar experiences
- the author and the type of journalism he does
- the professionalism he shown in calling out in his article the not-backed allegations other rivals made(for example of murder and sexual assault)
- the power dynamic that is usually in place between someone with enormous power and whealth, and a journalist that could be intimidated by being sued multiple times
Of course the amount/type of reasons needed to distrust someone is very personal, so we might need to "agree to disagree" on this
To be frank, While I tend to think that Dario has good intentions, I'm not so sure about his judgement. He's made a lot of claims that haven't panned out. I haven't felt that it was due to dishonesty, but more because of hyperbole.
The phrasing "Altman then denied having made the claim. “I didn’t even say that,” he said. “You just said that,” Daniela replied." is very close to the pattern I described above where someone interprets a claim as something different from what was actually said and refuses to back down. Furthermore this was prefaced with "As one person briefed on the exchange recalled" so it isn't even a first hand account. We don't know who the person doing the briefing was, but if it was one of the participants of the exchange, they would have been afforded the opportunity to reframe it to put themselves in a better light.
The second claim is potentially even more of a match for the example I gave regarding people misreading legal documentation. Was this a denial about the existence of words in a document, or was it a denial that the words represented the provision that was claimed. I have seen people do this, they take the existence of the words as proof of their interpretation and take dismissal of the interpretation as a claim that the words do not exist. I don't rightly know why people do this, but I have seen it happen. I suspect you could find an abundant supply of cases like this from the records of the worlds town council meetings.
It is difficult to assess the reliability of claims made by the current administration (understating it somewhat), but one of the things that was said about the Government negotiations with Anthropic was that he wanted a gate to some AI abilities in national security circumstance by requiring a personal phone call to Amodei to clear it. No sane government on earth would agree to something like that. It would be an invitation to providing a corporate interest a massive point of leverage in a time of crisis.
But again I am in a similar position with Amodei. I don't have any direct knowledge of the person so I will reserve judgement. I generally like the approach Anthropic is taking but the exposure I have had to the statements made by Amodei himself has given me pause. I would not condemn him either, but I also wouldn't place a lot of stock in what he says unless I see more to create a more complete view of his character.
You note amount of people interviewed and their very similar experiences but it's the nature of how those claims are similar that concerns me. So many of the claims seem to fall into the pattern that requires the person reporting the claim to judge the sole meaning of what was said. How many confirmed direct quotes have been confirmed to be untrue? I'm open to the evidence, perhaps this article will draw some out, but right now I see people convincing themselves of a pattern and then interpreting their own experiences in terms of that pattern.
The thing is, if you were to ask, I think Altman would agree that he shouldn't be in charge of the world's AI. I don't think any one person should, and I would treat anyone who claimed that they were the right person for that job with massive suspicion. To say that's where he sits is to buy into the premise that whoever is the head of OpenAI controls our future. OpenAI is but one of many enterprises working on this, there are a lot of people claiming they already have lost too much ground, but then there have been many predicting their imminent collapse, like a doomsday cult rolling forward the calendar whenever it doesn't happen.
Apologies, I didn't mean to highlight Amodei in those quotes, I just selected the sentence to have enough context but not be too long, it was a coincidence that they both started with Amodei. I'm not sure if those claims came from Amodei or not, nor I have any specific feeling about him.
> Furthermore this was prefaced with "As one person briefed on the exchange recalled" so it isn't even a first hand account
I'll admit I somehow missed that part, but we don't know how much of this event was in "Amodei's notes" and how much was from the "person briefed on the exchange"
> The phrasing "..." is very close to the pattern I described above where someone interprets a claim as something different
> The second claim is potentially even more of a match for the example I gave regarding people misreading legal documentation
I think our difference in point of view here lies on how much trust we put in the author, with what I seen so far I feel I have enough trust in the author to think he investigated these claims properly and made sure they weren't just misunderstandings, and that many of those checks he did weren't included in the article for any technical/legal reasons. Much more so reading some of his comments:
> As is always the case with incredibly precise and rigorously fact-checked reporting like this, where every word is chosen carefully (the initial closing meeting for this one was nearly eight hours long, with full deliberation about each sentence), there is more out there on that subject than is explicitly on the page.
> You try to reach a critical mass of detailed, rounded understanding of a central question, integrating the most meaningful perspectives, interrogating the weak points and blind spots, and backing up the assertions with documentary evidence or strong sourcing. Eventually, you reach a point where enough sources and materials are reliably triangulating toward the same truths.
> The fact-checking process at the New Yorker is exhaustive, and can span weeks. Every sentence, assertion, and piece of underlying sourcing get scrubbed by multiple independent pairs of eyes. This story had four fact-checkers working on it for the better part of a two week period, pulling very long hours.
As I said I'm happy to agree to disagree on this point.
> So many of the claims seem to fall into the pattern that requires the person reporting the claim to judge the sole meaning of what was said
I guess that's the nature of communications between humans. Even examples of written discussions seem contentious. The only type of claims I can think about that could be outside this category are the ones about written contracts, but it's understandable we don't have access to the actual contracts, and even if we did we couldn't really prove what was verbally agreed to be put in the contracts.
> To say that's where he sits is to buy into the premise that whoever is the head of OpenAI controls our future. OpenAI is but one of many enterprises working on this
This might start a whole new discussion, but I think being the CEO of one of the companies that produce state of the are models is enough to have a high concern. My worry is that he(or any other company) won't say "stop" if a new AI is found to be more powerful but have considerable negative impacts on society. As an example it doesn't matter who has the "strongest" atomic bomb, any country that has one is a potential treat to humanity and should have rigid controls in place.
I commented specifically on Altman because the article seems to suggest he's more power-greedy, persuasive, possibly deceptive, and with strong-leverages/contacts than the average person, or even the average CEO.
(edits: formatting)
I'm not pleased with the headline and the general framing that AI works. The plagiarism and IP theft aspects are entirely omitted. The widespread disillusion with AI is omitted.
On the positive side, the Kushner ad Abu Dhabi involvements (and threats from Kushner) deserve a wider audience.
My personal opinion is that "who should control AI" is the wrong question. In the current state, it is an IP laundering device and I wonder why publications fall silent on this. For example, the NYT has abandoned their crown witness Suchir Balaji who literally perished for his convictions (murder or not).
I would love to read your piece and pay you and new Yorker for it, but I am not interested in paying a subscription. If I could press a button and pay a reasonable one time license such as $3 or $5 for just this article, or better yet a few cents per paragraph as they load in, I wouldn't hesitate.
However I'm not going to pay for yet another subscription to access one article I'm interested in.
I'm sure you can't do anything about this, but I just wanted you to know.
You deserve to be compensated for great journalism. In this case, unfortunately, I won't read it and you won't earn income from me.
You're already paying for your library with your tax dollars. If you don't use it, you may lose it, but you will certainly lose out by subsidizing bums, vagrants, and other families who use the library to their heart's content.
The public library also features lots of streaming and CD music, videos, and video games, that you can freely check out without any cost. In fact, my local library staff told me that they've abolished overdue fees. Libby and the digital apps will automatically renew or return materials. My physical books even got auto-renewed three times before I needed to manually do it, or bring them back into the building.
> “Investors are, like, I need to know you’re gonna stick with this when times get hard,”
Should be:
> “Investors are like, I need to know you’re gonna stick with this when times get hard,”
In "investors were like, you need to grow", you're semi-quoting someone, and can't take it out: "investors were you need to grow".
Or Mr Farrow can you post some evidence somewhere we can see?