The argument doesn't hinge on whether OpenAI is actually open. Rather it seems to have to do with the name being insufficiently distinguishable from a generic term ("open AI"). I think it's a bizarre ruling given that everyone already knows what OpenAI is.
Trademark law isn't about what "everyone already knows". It's about whether a given mark meets the criteria for legal protection in a give context. So if say an foss ML project described what they do as "open AI" the company known as OpenAI would have a right to defend the mark. This is saying they could not.
> So if say an foss ML project described what they do as "open AI" the company known as OpenAI would have a right to defend the mark. This is saying they could not.
Well if that's all that's at stake here, it seems very reasonable.
Everyone on HN knows what OpenAI is, but there are tons of people who use ChatGPT and either don’t know OpenAI or don’t know the distinction between OpenAI (the company) and OpenAI (the conjunction of two words)
The linguistic gymnastics required when talking about OpenAI vs ChatGPT and Anthropic vs Claude is difficult when you're giving talk about them. At least Google vs Gemini is a little clearer.
I mean, I get the rationale Company vs. Product, but most people know the product. As in "I used ChatGPT". But if you ask who OpenAI is, they'll have no clue.
ChatGPT is in someways nicer... because their models are GPT-5.3, GPT-5.4, etc...
But when you're trying to explain that the Anthropic models are called "Opus" or "Sonnet" or "Haiku" or "Fable", but you use them in "Claude", it gets confusing quickly.
> But when you're trying to explain that the Anthropic models are called "Opus" or "Sonnet" or "Haiku" or "Fable", but you use them in "Claude", it gets confusing quickly.
"Why is your French coworker helping you write literature? I thought you worked in software"
I mentioned this thread to my wife, and she told me that I was the only person she had ever heard mention the name "OpenAI", and she wouldn't have any clue who they were if she saw their name somewhere if I hadn't told her in the past.
If the goal of a trademark is to get recognized then its futile given OpenAI is already popular. If the goal is to prevent others from using the term which is so generic then it does makes sense to not allow the common keywords being hijacked.
On a side note, the AI models from the company are not even open, one can go as far as banning it as inappropriate marketing (Product not matching the description).
I thought the trademark is to prevent costumers from accidentally buying “open AI” from some company other than openAI, while thinking they’re buying from openAI.
Yes, that is what a trademark is for, in general. In this case, the court ruled that the term "open AI" is too generic to qualify for that protection exactly because it is a purely descriptive term that could refer to any "freely available" model.
They haven't prevented that. They have prevented trademarking the terms, thus other people whose AI offerings are Open are in fact allowed to describe their products as an Open AI, I presume they are not allowed to describe their products as being OpenAI however as that would create consumer confusion.
Furthermore they have not said anything about [adjective] being non trademarkable, they have said that you shouldn't be able to trademark things that have specific meaning in your industry, as Open has some specific meaning in the software industry.
Thus you would probably be allowed to name your things [big] +[proudct/service provided] or in this case bigAI because big does not really imply a specific desirable quality in the Software industry.
Now before you start talking of how you can see blah blah how big would be useful blah blah, as is the tradition whenever programmers encounter a legal decision that they do not agree with, it just ain't gonna work. I guess though I cannot prevent the inevitable, but nobody in IT says does it have the technical quality of "bigness" before purchasing, but they do about the quality of "Openness", so obviously some adjectives would be untrademarkable in this context, if you named your AI SecureAI probably no go, If you named your AI UglyAssAI probably fine.
This seems a lot more sustainable than allowing me to trademark a tire company called "WinterTire" and enabling me to sue any other tire company that tries to capitalise on my trademark.
(And if WinterTire Co was anything like OpenAI, it'd be focused on making summer tires)
Preventing the hijacking and privatization of short phrases and language in general is actually an excellent thing. I applaud this decision, and wish for the rules to become even tighter.
This ruling is very early in the process and doesn't prevent anything - but given the more likely bad path outcome of this case for OpenAI it wouldn't disallow their usage of OpenAI but instead prevent them from shutting down competitors that claim to offer an open AI.
I seem to remember the company behind either Monster Cables, or Monster energy drinks, going after anyone that used the word “monster,” even in casual context.
Actually I have grabbed a couple of USB-C Monster-branded cables at the hardware store, and they work great as replacement charging cables for my devices (Chromebook and Pixel phoned) for rapid charging at 6 feet rather than 3.
The most ludicrous case of trademark issues I've seen to date has been Apple suing Prepear because they had a pear as their logo but Monster suing someone because of the term monster is not far from falling equally as far from the tree of wisdom and common sense.
This really is the crux of the image, and now legal, issues. OpenAI hovered up tons of money and IP under the claim that they were doing this for the public good. Now they’ve essentially admitted that was all bullshit and that they want to sell the distillation of human created knowledge and content for a fee. It’s certainly bullshit to call that bait and switch “open”
You know, if it was someone offering a truly open (weights + training data) and available model running on consumer hardware in a privacy sandbox, I would welcome that "harm".
> The EUIPO found that the word "open" would be understood by the relevant public as meaning freely accessible, while the combination with "AI" (artificial intelligence) would be interpreted as referring to products based on openly accessible artificial intelligence.
More pertinently "the term is purely descriptive and therefore lacks the distinctiveness required for trademark protection." I.e. the problem isn't that OpenAI's products don't match their description, but that trademarking it would unduly prevent others from describing their openly accessible artificial intelligence as "open AI."
>More pertinently "the term is purely descriptive and therefore lacks the distinctiveness required for trademark protection."
As I wrote in my other comment, "open systems" also can be purely descriptive and yet Open Systems seems to be a valid trademark in Europe.
I'm not defending OpenAI. I'm just confused that the rules for allowing trademarks for ordinary words and phrases don't look consistent at the surface level.
From the background information provided in the court decision:
"as regards the earlier similar registrations relied on by the applicant, the Board of Appeal recalled that those registrations did not represent current practice and case-law and that the legality of the decisions of the Boards of Appeal must be assessed solely on the basis of Regulation 2017/1001 and not on the basis of a previous administrative practice."
> I'm just confused that the rules for allowing trademarks for ordinary words and phrases don't look consistent at the surface level
Precedence in court decisions is weighted more in some places than others. US courts, from my pedestrian observation, are all into "well there is a precedent so we'll follow that". But other countries don't necessarily do that, their courts might be more eager to disregard it and do whatever they feel is appropriate in that situation. You can see it in how OpenAI reacted "The company also cited comparable trademark registrations previously granted by the EUIPO and registrations in more than 30 other countries, including the United Kingdom and Singapore" (well precedents say this and that...)
Even the US does change policies every once in a while. This is a case where Open Systems was last challenged under the old rule set when it was acceptably descriptive. Since that point standards have changed and, presumably, Open Systems might now be exposed to a similar mark challenge.
This sort of rejection is quite common. You can overcome the objection by submitting evidence that your mark has acquired distinctiveness through use. So there will be plenty of trademarks that seem generic/descriptive that have gone through that process.
> I originally thought that’s what was being rejected here. It’s not. The court only ruled that “OpenAI” is descriptive. Separately, OpenAI have also submitted evidence of acquired distinctiveness, which will be decided in due course by the trademark office.
My gut feeling is on them loosing that one.
Stop most people on the street and ask them about "OpenAI". I suspect your average Joe would say "Who ?".
Continue the discussion with "have you heard of ChatGPT" and I suspect the answer would mostly be "Yes, I have".
OpenAI are applying for "OpenAI" in this trademark application. The fact they have a widely known product called "ChatGPT" which they actively market under the name "ChatGPT" does not and should not come into the thought process of any appeals judge.
The fact that Open Systems hold a trademark on "Open Systems" is less a signal that OpenAI should be allowed to hold "Open AI", and more a sign that Open Systems should start considering changing their name because they'd lose their trademark if someone disputed it.
And to think that this could have all been avoided if they'd just renamed themselves something more appropriate after they decided to focus fully on developing closed models for profit.
The purpose of the company is also rarely defined by it's name. Brands aren't inherently descriptive and it's normal for companies to evolve over time.
This feels like a slight misstep that could result in consumer harm. The name is incredibly vague, without doubt, but to claim "OpenAI" doesn't evoke a very specific company at this point in the minds of consumers seems myopic.
But people actively searching for AI products who are perhaps a little less technically inclined might. And if they stumble upon a platform that by all accounts seems to be affiliated with OpenAI, that could be problematic, especially with the level of trust people seem to be comfortable handing to LLMs.
If I search for "Open AI" on google right now the first search results are openai.com, chatgpt.com and the OpenAI wikipedia page. None of which are open AI.
But that's on OpenAI for selecting that company name. EU trademark law hasn't changed, this was always going to be a problematic trademark if challenged
That appears to be saying you cannot make a trademark of something that is well known to be used by someone else, which is very different from needing the new trademark you are trying to register to be well known.
The only problem I see here is the name doesn't reflect the reality. Time to put something in place that tells them to rebrand and continuously charges them for fraudulent misrepresentation or something until they do.
Key difference between the trademark systems here: in the EU system you don’t get a trademark by trading with a specific name and it then being recognized. It’s the other way around: the name must be unique, not confusing, and highly specific. It’s actually irrelevant whether a product exists or is traded at all.
Having gone through the process and gotten both approvals and rejections, the line is pretty clear.
"Apples in the Sky" is a valid trademark only because apples in the sky do not exist. If there was a strange meteorobiological event where such fruit started to grow in the clouds, this would no longer be a valid trademark for someone to create, because it would be descriptive of a category of things in the real world.
I don't like the idea of trademarks that are so well-known that they apply in all fields. That doesn't serve the public. Companies that well-known can easily apply for a trademark across multiple fields. The trademark owners that would really need that kind of help are the smaller ones that aren't eligible for that, and must specifically register in various fields of industry.
IP law needs severe reform no matter which jurisdiction you're in (since the majority of the world signed the Berne Convention, the same reforms are needed everywhere)
But what's the alternative for the trademarks that are truly universally known? Does it serve the public for me to be able to sell unauthorized Coca-Cola brand products, or more likely, the trademark owner selling a token item in each category to maintain the right to prevent others from using it?
The story about the ruling really doesn't explain why another company called OpenText that's been around since 1991 and has a valid trademark registration in EU but OpenAI would be invalid. OpenText also has its Europe headquarters in Germany: https://www.opentext.com/about/office-locations
Any legal guesses as to why those 2 companies are treated differently with regards to the very generic words : "open", "text", "AI" ?
> Any legal guesses as to why those 2 companies are treated differently with regards to the very generic words : "open", "text", "AI" ?
The legal situation may have changed since 1991. For example: The ruling refers to "Regulation 2017/1001", which, as the name suggests, only came into force in 2017.
Open AI has an independent descriptive meaning as composite term. You would practically trademark a whole class of products, not only a brand name.
In contrast, open text is not descriptive in the sense of being a category of things. Therefore there is no risk that competitors would run into trademark issues by just describing their products.
Also, trademark decisions are always contextual to their time. Today’s meaning of ‘open’ in the context of software and data was not even coined in 1991, at that time people used ‘free software’ as term. Today I am not sure if ‘open text’ could still be trademarked.
>Open AI has an independent descriptive meaning as composite term.
See my edit. "Open Systems" also had an independent descriptive meaning. The phrase "open systems" was a very common generic phrase in 1990s when companies talking about POSIX compliance was a big deal. (E.g. Microsoft touted POSIX in Windows NT.)
As much as I hate OpenAI for hijacking the term "open", and I love the idea of OpenAI losing, I am not sure if I agree with it.
Trademarks are first intended to protect consumers, so that if it says Coca Cola, then the Coca Cola company made it, for the better of for the worse, but at least you know.
OpenAI is already a well known name in Europe, and when I see OpenAI on a product, I expect it to be a product of that company. It doesn't mean I will want to use it, I may even want to avoid it, but I don't want it to be from someone else. By denying that trademark, anyone could call their product OpenAI, and I don't think that situation would benefit the consumer.
No, they are very much not. That's the supposed benefit of patents and/or copyright. Trademarks are for ensuring there is no consumer confusion about which company they are trading with.
As explained in the judgement, being well-known is irrelevant. This is not about consumer protection, but about brand protection.
You are right that the decision has the potential to confuse consumers. However, that is on Open AI, they should have consulted trademark lawyers earlier, and should have rebranded after shifting from open AI to commercial AI.
At least the GL part isn't something that has a widely understood meaning and I don't think any of the competing APIs are generally refered to as Graphics Libraries either.
This means light green to all EU tech companies using OpenAI name in their products!
Even though can´t say for sure if is good or bad for a company doing that.
The court only ruled that "OpenAI" is descriptive. It did not rule that it can never be registered.
A descriptive trademark can still be registered with evidence that the mark has become distinctive through use -- that is, that "OpenAI" is known to the general public to refer to their company, rather than simply describing a type of AI.
It appears OpenAI have already made that submission separately:
> The Board of Appeal also stated that, once the contested decision has become final, the proceedings will resume for the purposes of examining the applicant’s alternative claim based on Article 7(3) of Regulation 2017/1001, relating to distinctive character acquired through use.
120 comments
[ 3.1 ms ] story [ 60.4 ms ] threadWell if that's all that's at stake here, it seems very reasonable.
I mean, I get the rationale Company vs. Product, but most people know the product. As in "I used ChatGPT". But if you ask who OpenAI is, they'll have no clue.
ChatGPT is in someways nicer... because their models are GPT-5.3, GPT-5.4, etc...
But when you're trying to explain that the Anthropic models are called "Opus" or "Sonnet" or "Haiku" or "Fable", but you use them in "Claude", it gets confusing quickly.
"Why is your French coworker helping you write literature? I thought you worked in software"
He began the conversation by saying: “so as far as I know there’s two kinds of ai, open ai and closed ai…”
This speaks to your point, the people are confused.
If it has Open in the name it's something to do with open source and "AI" right? :)
As such "everyone knows them" isn't a reason to allow a registration. It would just mean that blocking the trademark has no practical effect
On a side note, the AI models from the company are not even open, one can go as far as banning it as inappropriate marketing (Product not matching the description).
Caterpillar, Apple, Kellogg, etc really don’t have anything to do with the underlying product but neither do people’s names.
Nintendo could have named itself after playing cards, but that wouldn’t have kept up with its current business model.
Furthermore they have not said anything about [adjective] being non trademarkable, they have said that you shouldn't be able to trademark things that have specific meaning in your industry, as Open has some specific meaning in the software industry.
Thus you would probably be allowed to name your things [big] +[proudct/service provided] or in this case bigAI because big does not really imply a specific desirable quality in the Software industry.
Now before you start talking of how you can see blah blah how big would be useful blah blah, as is the tradition whenever programmers encounter a legal decision that they do not agree with, it just ain't gonna work. I guess though I cannot prevent the inevitable, but nobody in IT says does it have the technical quality of "bigness" before purchasing, but they do about the quality of "Openness", so obviously some adjectives would be untrademarkable in this context, if you named your AI SecureAI probably no go, If you named your AI UglyAssAI probably fine.
(And if WinterTire Co was anything like OpenAI, it'd be focused on making summer tires)
https://www.worldtrademarkreview.com/article/monster-energy-... https://techraptor.net/gaming/news/gods-and-monsters-started... https://www.bbc.com/news/uk-england-berkshire-46369442 https://www.thegamer.com/monster-energy-goes-after-glowstick... https://www.koreaboo.com/news/yg-entertainment-wins-trademar... https://www.gamesradar.com/monster-energy-has-even-gone-afte...
and many more.
https://www.engadget.com/2008-03-03-audiophiles-cant-tell-th...
Can't help to notice that in this case lawyers tell companies to buy more lawyer time.
As I wrote in my other comment, "open systems" also can be purely descriptive and yet Open Systems seems to be a valid trademark in Europe.
I'm not defending OpenAI. I'm just confused that the rules for allowing trademarks for ordinary words and phrases don't look consistent at the surface level.
"as regards the earlier similar registrations relied on by the applicant, the Board of Appeal recalled that those registrations did not represent current practice and case-law and that the legality of the decisions of the Boards of Appeal must be assessed solely on the basis of Regulation 2017/1001 and not on the basis of a previous administrative practice."
Precedence in court decisions is weighted more in some places than others. US courts, from my pedestrian observation, are all into "well there is a precedent so we'll follow that". But other countries don't necessarily do that, their courts might be more eager to disregard it and do whatever they feel is appropriate in that situation. You can see it in how OpenAI reacted "The company also cited comparable trademark registrations previously granted by the EUIPO and registrations in more than 30 other countries, including the United Kingdom and Singapore" (well precedents say this and that...)
My gut feeling is on them loosing that one.
Stop most people on the street and ask them about "OpenAI". I suspect your average Joe would say "Who ?".
Continue the discussion with "have you heard of ChatGPT" and I suspect the answer would mostly be "Yes, I have".
OpenAI are applying for "OpenAI" in this trademark application. The fact they have a widely known product called "ChatGPT" which they actively market under the name "ChatGPT" does not and should not come into the thought process of any appeals judge.
Take, for example, OpenText.
Nothing about it is free, or open source code, or in any other sense publicly accessible. [2]
I don't know if they operate in the EU, but I presume any application for a trademark would be rejected.
[1] https://en.wikipedia.org/wiki/OpenText
[2] It's "Open" as it turns paper documents into more easily accessible digital formats. But not public in any way.
[1] It originally did in 1833. It doesn't today.
It would be more honest to their customers and better show who they are and what they stand for.
They still have the trademark on their logo
Although courts make mad decisions sometimes...
Having gone through the process and gotten both approvals and rejections, the line is pretty clear.
Here are the 13 valid trademarks in France containing the word "apple" in the same category as fruit: https://data.inpi.fr/search?advancedSearch=%257B%2522checkbo...
None of them are descriptive of the actual fruit.
"Apples in the Sky" is a valid trademark only because apples in the sky do not exist. If there was a strange meteorobiological event where such fruit started to grow in the clouds, this would no longer be a valid trademark for someone to create, because it would be descriptive of a category of things in the real world.
IP law needs severe reform no matter which jurisdiction you're in (since the majority of the world signed the Berne Convention, the same reforms are needed everywhere)
The panel ruled in our favor, that their OPENSPACE trademark is probably invalid because it is descriptive.
https://domainnamewire.com/2026/04/08/u-s-defense-contractor...
Tangentially www.openspace.ai also exists.
Any legal guesses as to why those 2 companies are treated differently with regards to the very generic words : "open", "text", "AI" ?
The legal situation may have changed since 1991. For example: The ruling refers to "Regulation 2017/1001", which, as the name suggests, only came into force in 2017.
In contrast, open text is not descriptive in the sense of being a category of things. Therefore there is no risk that competitors would run into trademark issues by just describing their products.
Also, trademark decisions are always contextual to their time. Today’s meaning of ‘open’ in the context of software and data was not even coined in 1991, at that time people used ‘free software’ as term. Today I am not sure if ‘open text’ could still be trademarked.
See my edit. "Open Systems" also had an independent descriptive meaning. The phrase "open systems" was a very common generic phrase in 1990s when companies talking about POSIX compliance was a big deal. (E.g. Microsoft touted POSIX in Windows NT.)
Trademarks are first intended to protect consumers, so that if it says Coca Cola, then the Coca Cola company made it, for the better of for the worse, but at least you know.
OpenAI is already a well known name in Europe, and when I see OpenAI on a product, I expect it to be a product of that company. It doesn't mean I will want to use it, I may even want to avoid it, but I don't want it to be from someone else. By denying that trademark, anyone could call their product OpenAI, and I don't think that situation would benefit the consumer.
Huh? I thought they're intended to protect "innovation".
Trademarks don't prevent you from copying anything, they only prevent you from being misleading regarding the origin.
You are right that the decision has the potential to confuse consumers. However, that is on Open AI, they should have consulted trademark lawyers earlier, and should have rebranded after shifting from open AI to commercial AI.
I am also not discussing about who is at fault, I agree that it is on OpenAI.
I just don't want, say, some company that is even shadier than OpenAI to launch an OpenAI branded protect with the intention to mislead people.
Maybe grant OpenAI the trademark, but do not allow them to use it on products that are not actually open, but I guess it is legally problematic.
Duh. The open in OpenAI isn't supposed to mean open. We've all been complaining about nothing.
What does OpenText mean? [1]
They do a lot of classified work for the US government.
[1] https://en.wikipedia.org/wiki/OpenText
The court only ruled that "OpenAI" is descriptive. It did not rule that it can never be registered.
A descriptive trademark can still be registered with evidence that the mark has become distinctive through use -- that is, that "OpenAI" is known to the general public to refer to their company, rather than simply describing a type of AI.
It appears OpenAI have already made that submission separately:
> The Board of Appeal also stated that, once the contested decision has become final, the proceedings will resume for the purposes of examining the applicant’s alternative claim based on Article 7(3) of Regulation 2017/1001, relating to distinctive character acquired through use.
"Closed AI" would be a better name