remember a corporation generally is an object owned by some people. Do you trust "unspecified future group of people" with your privacy? You can't. Best we can do is understand the information architecture and act accordingly.
"How dare the New York Times demand access to our vault of everything-we-keep to figure out if we're a bunch of lying asses. We must resist them in the name of user privacy! Signed, the people who have scraped literally everything to incorporate it into the products we make."
OpenAI may be trying to paint themselves as the goody-two-shoes here, but they're not.
Cynicism aside, this seems like an attempt to prune back a potentially excessive legal discovery demand by appealing to public opinion.
The New York Times is demanding that we turn over 20 million of your private
ChatGPT conversations. They claim they might find examples of you using
ChatGPT to try to get around their paywall.
I wouldn't want to make it out like I think OpenAI is the good guy here. I don't.
But conversations people thought they were having with OpenAI in private are now going to be scoured by the New York Times' lawyers. I'm aware of the third party doctrine and that if you put something online it can never be actually private. But I think this also runs counter to people's expectations when they're using the product.
In copyright cases, typically you need to show some kind of harm. This case is unusual because the New York Times can't point to any harm, so they have to trawl through private conversations OpenAI's customers have had with their service to see if they can find any.
Man, maybe I'm getting old and jaded, but it's not often that I read a post that literally makes my skin crawl.
This is so transparently icky. "Oh woe is us! We're being sued and we're looking out for YOU the user, who is definitely not the product. We are just a 'lil 'ol (near) trillion-dollar business trying to protect you!"
Come ON.
Look I don't actually know who's in the right in the OAI vs. NYT dispute, and frankly I personally lean more toward the side the says that you are allowed to train models on the world's information as long as you consume it legally and don't violate copyright.
But this transparent attempt to get user sympathy under insanely disingenuous pretenses is just absurd.
This screams just as genuine as Google saying anything about Privacy.
Both companies are clearly wrong here. There is a small part of me that kinda wants openai to loose this, just so maybe it will be a wake up call to people putting in way too personal of information into these services? Am I too hopeful here that people will learn anything...
Fundamentally I agree with what they are saying though, just don't find it genuine in the slightest coming from them.
If Donald Trump used this OpenAI product to-- who knows-- brainstorm Truth Social content, and his chats were produced to the NYT as well as its consultants and lawyers, who would believe Mr. Trump's content remained secure, confidential and protected from misuse against his wishes?
That's simply a function of the fact it's a controversial news organization running a dragnet on private communications to a technology platform.
Almost every comment (five) so far is against this: 'An incredibly cynical attempt at spin', 'How dare the New York Times demand access to our vault of everything-we-keep to figure out if we're a bunch of lying asses', etc.
In direct contrast: I fully agree with OpenAI here. We can have a more nuanced opinion than 'piracy to train AI is bad therefore refusing to share chats is bad', which sounds absurd but is genuinely how one of the other comments follows logic.
Privacy is paramount. People _trust_ that their chats are private: they ask sensitive questions, ones to do with intensely personal or private or confidential things. For that to be broken -- for a company to force users to have their private data accessed -- is vile.
The tech community has largely stood against this kind of thing when it's been invasive scanning of private messages, tracking user data, etc. I hope we can collectively be better (I'm using ethical terms for a reason) than the other replies show. We don't have to support OpenAI's actions in order to oppose the NYT's actions.
So why aren’t they offering for an independent auditor to come into OpenAI and inspect their data (without taking it outside of OpenAI’s systems)?
Probably because they have a lot to hide, a lot to lose, and no interest in fair play.
Theoretically, they could prove their tools aren’t being used to doing anything wrong but practically, we all know they can’t because they are actually in the wrong (in both the moral and, IMO though IANAL, the legal sense). They know it, we know it, the only problem is breaking the ridiculous walled garden that stops the courts from ‘knowing’ it.
I keep asking ChatGPT how to get NYT articles for free and then add lots of vulgar murderous things about their lawyers in the same message. It’s a private thought to an AI, so the attorneys can’t complain, right?
Why should OpenAI keep those conversations in the first point? (of course the answer is obvious) If they didn't keep them, they wouldn't have anything to hand over, and they would have protected users' privacy MUCH better. This is just as good as Facebook or Google care about their users' privacy.
Gmail is an Electronic Communication Service as defined in 18 U.S.C § 2510, meaning its contents are protected under the Stored Communications Act (18 U.S.C. Chapter 121 §§ 2701–2713).
Communications with an AI system do not involve a human so are not protected by ECPA or the SCA and get less protection. This is controversial and some people have called on ECPA/SCA to be extended to cover AI services. That means a warrant would be necessary to get your OpenAI history, not just a subpoena.
Wondering if anyone here has a good answer to this:
what protection does user data typically have during legal discovery in a civil suit like this where the defendant is a service provider but relevant evidence is likely present in user data?
Does a judge have to weigh a users' expectation of privacy against the request? Do terms of service come into play here (who actually owns the data? what privacy guarantees does the company make?).
I'm assuming in this case that the request itself isn't overly broad and seems like a legitimate use of the discovery process.
I fully believe that OpenAI is essentially stealing the work of others by training their models on it without permission. However, giving a corporation infamous for promoting authoritarianism full access to millions of private conversations is not the answer.
OpenAI is right here. The NYT needs to prove their case another way.
100 comments
[ 3.1 ms ] story [ 88.8 ms ] thread-- openai
remember a corporation generally is an object owned by some people. Do you trust "unspecified future group of people" with your privacy? You can't. Best we can do is understand the information architecture and act accordingly.
OpenAI may be trying to paint themselves as the goody-two-shoes here, but they're not.
But conversations people thought they were having with OpenAI in private are now going to be scoured by the New York Times' lawyers. I'm aware of the third party doctrine and that if you put something online it can never be actually private. But I think this also runs counter to people's expectations when they're using the product.
In copyright cases, typically you need to show some kind of harm. This case is unusual because the New York Times can't point to any harm, so they have to trawl through private conversations OpenAI's customers have had with their service to see if they can find any.
It's quite literally a fishing expedition.
This is so transparently icky. "Oh woe is us! We're being sued and we're looking out for YOU the user, who is definitely not the product. We are just a 'lil 'ol (near) trillion-dollar business trying to protect you!"
Come ON.
Look I don't actually know who's in the right in the OAI vs. NYT dispute, and frankly I personally lean more toward the side the says that you are allowed to train models on the world's information as long as you consume it legally and don't violate copyright.
But this transparent attempt to get user sympathy under insanely disingenuous pretenses is just absurd.
Both companies are clearly wrong here. There is a small part of me that kinda wants openai to loose this, just so maybe it will be a wake up call to people putting in way too personal of information into these services? Am I too hopeful here that people will learn anything...
Fundamentally I agree with what they are saying though, just don't find it genuine in the slightest coming from them.
That's simply a function of the fact it's a controversial news organization running a dragnet on private communications to a technology platform.
"Great cases, like hard cases, make bad law."
And what if they for example find evidence of X other thing such as:
1. Something useful for a story, maybe they follow up in parallel. Know who to interview and what to ask?
2. A crime.
3. An ongoing crime.
4. Something else they can sue someone else for.
5. Top secret information
In direct contrast: I fully agree with OpenAI here. We can have a more nuanced opinion than 'piracy to train AI is bad therefore refusing to share chats is bad', which sounds absurd but is genuinely how one of the other comments follows logic.
Privacy is paramount. People _trust_ that their chats are private: they ask sensitive questions, ones to do with intensely personal or private or confidential things. For that to be broken -- for a company to force users to have their private data accessed -- is vile.
The tech community has largely stood against this kind of thing when it's been invasive scanning of private messages, tracking user data, etc. I hope we can collectively be better (I'm using ethical terms for a reason) than the other replies show. We don't have to support OpenAI's actions in order to oppose the NYT's actions.
Probably because they have a lot to hide, a lot to lose, and no interest in fair play.
Theoretically, they could prove their tools aren’t being used to doing anything wrong but practically, we all know they can’t because they are actually in the wrong (in both the moral and, IMO though IANAL, the legal sense). They know it, we know it, the only problem is breaking the ridiculous walled garden that stops the courts from ‘knowing’ it.
https://www.schneier.com/blog/archives/2025/06/what-llms-kno...
At some point they'll monetize these dossiers.
Communications with an AI system do not involve a human so are not protected by ECPA or the SCA and get less protection. This is controversial and some people have called on ECPA/SCA to be extended to cover AI services. That means a warrant would be necessary to get your OpenAI history, not just a subpoena.
what protection does user data typically have during legal discovery in a civil suit like this where the defendant is a service provider but relevant evidence is likely present in user data?
Does a judge have to weigh a users' expectation of privacy against the request? Do terms of service come into play here (who actually owns the data? what privacy guarantees does the company make?).
I'm assuming in this case that the request itself isn't overly broad and seems like a legitimate use of the discovery process.
I think I have enough with the first sentence, no need to read more. The narration is clear, we are the brain and no one can stop us.
OpenAI is right here. The NYT needs to prove their case another way.