AI cannot be the inventor of a patent, appeals court rules
bbc.com
bbc.com
The scope of patents in software and technology is already absurd, both because 12 years is an eternity for how quickly that industry moves, and because increasingly trivial "inventions" have been let through as the ability for patent officials to interpret complicated digital innovations according to patent law and policy has increasingly fallen behind the field. This is already a serious problem, but currently still takes expensive lawyers to take advantage of. Even if we make legal precedent that says that AI can't own patents, the advent of better purpose-specific text generation will quickly put patent trolling in the hands of laypeople
Personally, I think the DMCA was a horrible mistake. The provisions it added to copyright created a ton of horrible precedent that led to the draconian control private companies exert over the lives of billions of people worldwide, justified in law by their intellectual property concerns. I believe that if Intellectual Property ever served a legitimate purpose, it has now run its course and needs to be dismantled. I hope that this new challenge moves us more in this direction, rather than some new awful legislative band-aid that attempts to preserve the status quo by destroying more of our rights
1. IP assignment is overwhelmingly not retained by the creator of any work covered by IP. The vast overwhelming majority of engineers, for example, have in their contract of employment assigned all rights to any IP they generate in their work to the company they work for. The same is true of artists who contract with distributors
2. Independent creators often have no ability to successfully defend their IP claims, because said claims require expensive litigation, the cases are never straightforward enough for a summary judgement, and the individual nature of the claims mean that a class action lawsuit is nearly always infeasible
3. The way that IP law has been interpreted, especially under the DMCA, is incredibly abusable. The early history of the DMCA saw massive lawsuits filed by corporations against individuals based on obviously ridiculous numbers calculated as "potential lost sales" using flimsy evidence. A whole new business model of tech patent trolling has created a cottage industry that makes as many spurious claims as it can in order to leech money from businesses without producing anything of value. The current reading of the DMCA stymies users' attempts to repair their devices or even protect their privacy, because they can be criminally liable for attempts to modify devices that can be argued to be partially acting as DRM mechanisms, sometimes on the code running on the device itself.
When we talk about the value of a law or policy, the intent of the law is at best a diagnostic tool for an untested new kind of attempt at something. At the end of the day, it doesn't matter what a policy or law intends, it matters what the effects of the law are when interpreted and enforced. Intellectual Property as a legal paradigm generally is malfunctioning in a lot of places that harm people, harm society, and fail to uphold its promise. I think that the concept of ownership over ideas as a whole is at this phase in humman history doing much, much more harm than good, and needs to be eroded and eventually abolished from law
As for how creative people can make money off of their work in an environment without IP, there's actually a pretty clear answer for this that is not even my observation. Most creative work right now is paid for at production, not based on IP rights. The engineers working at a company are paid for their time and labor, and sign over the IP they produce. Independent artists work on commissions, and at a larger scale, crowdfund their projects. It is only the occasionally very established and wealthy artist or inventor that can really take advantage of IP laws, and their beneficiaries are as it stands mostly large corporations, treating IP as effectively a dragon hoard of speculative assets they can monetize through litigation and threats thereof. Most of the effect of this is stifling, not encouraging, independent innovation
Also, international agreements like the WTO have put the entire globe in a worse situation with regards to COVID vaccines - which need rapid, global distribution to be effective - because of patent rights over those vaccines. While many vaccines have resulted from a large amount of public funding, the IP controlling their production is largely in the hands of private corporations which view non-affiliated production of vaccines as cutting into their bottom line. They not only refuse to share the know-how to ramp up production with organizations trying to build up this capability worldwide, but have threatened to actively shut down such efforts if they happen anyway. This is obviously terrible for humanity as a whole, as we are currently in a situation where existing production and distribution is inadequate to keep ahead of variant evolution
What justifies your stronger claim that IP itself should be done away with?
1. We understand that ideas build on each other. It has always worked this way, and it is overwhelmingly the way humans get things done. Every scientific discovery stands on the shoulders of giants. Every story or song is influenced by ones that came before it. Creativity is valuable, but to define the entirety of a creative endeavor as undeniably the effort of a single person is insane. Because creativity builds on other creativity, any legal concept that tries to parcel out unique contributions of individuals or groups and impose a tariff on them is likely to stifle innovation, because any time it actually comes to matter, it can only create an unnatural block on a creative effort
2. Most people who have worked in a creative field understand that a concept or idea is seldom of much value, and must be executed in some way to be meaningful. Various concepts of intellectual property even try to capture this, e.g. by granting rights over performances or recordings specifically. The best arguments I've heard for "intellectual property" abstractly are as an interim solution for allowing collaborations to produce economic value for the person who didn't control the distribution, but this essentially can't function if we treat these rights as "property", because property entails the right to transfer this property, creating a mechanism to effectively void these rights.
3. Attribution is a valuable ethical principle, and giving people due credit for their ideas is an important part of creative endeavors from science to music, but property rights are simply not the correct framework to legislate this with. False claims of attribution can be fraud or sometimes defamation, and attribution can be proven by the well-understood mechanism of demonstrable prior art. Proof of prior art is more technologically feasible to ensure in an era where we have cryptographic signatures to uniquely prove authenticity and the ability to easily publish something with a timestamp to an archive
I don't think that's fair. I'm responding to a claim that IP is "wrong". The parent poster purported to demonstrate why this is so, and had not done this. The burden of proof is not on me.
You spoke of principles, but I think you got them backwards, and thus sought to clarify what the actual underlying principles are.
I agree with you that the law is dysfunctional in many ways. I especially agree with you that a law should be judged on the outcomes it creates rather than the intent of the policy that motivates it. But I’d suggest the appropriate solution is to improve the law, rather than abolishing IP rights.
Without IP rights, we’d just return to industrial trade secrets, which would only really benefit big business. You would see much less investment in areas where trade secrets are not possible (like most pharmaceutical research for instance). A lot of people already think current-day DRM controls are anti-consumer, but how does a cinema-only movie release sound?
As for art and media, the DMCA had a real problem to solve: The business models of the massive distribution conglomerates that had arisen when publishing and copying information was expensive were infeasible in the face of digital computers that could copy any information with high fidelity and distribute it over the internet. This business model should have collapsed in the face of this technology. The market would have been forced to adjust to some new model, e.g. crowdfunding (as has, again, already happened in even relatively large endeavors considered "independent" because they are not backed by a massive conglomerate). Instead, the market's status quo was prioritized. What needed to happen was for the market to equillibrate to the new reality of these technologies, and it's insane to view the "piracy" fights of the 90s as anything other than that process. DRM controls were, as a technology, a zero-sum game played by distributors against consumers, and some of them should - by any reasonable interpretation of laws against malicious access to a device owned by someone else - be illegal. Criminalization of circumventing DRM is a horrible monster of a law that puts a heavy criminal thumb on the scale in this conflict, prevented the market from equilibrating in favor of consumers or even new players trying new business models in any way, and cemented the dominance of existing players by fiat, giving them the backing of a police state to effectively enforce their business models as law
This very obviously did occur, though perhaps not as quickly as it might have otherwise. The market of CD and DVD sales has absolutely collapsed, and streaming services have taken over, regardless of that fact that this was initially resisted by the industry. But this new paradigm, which offers a much better product to consumers, is only possible because of IP rights. Piracy is no longer the existential threat to this industry that it once was, because legally licensed content has been able to offer a superior user experience (in most cases). But if I was allowed to set up a Netflix clone, host anything I wanted on it, and undercut their prices, then capital allocation to content creation would evaporate immediately. The thing you’re trying to promote access to simply doesn’t exist without these protections, and those protections don’t prevent anybody from crowdsourcing something if they want to.
> In practice, trade secrets don't usually stand up to sophisticated reverse-engineering efforts, and patents frequently and consistently fail to disclose implementation details that are important to how something works.
The ubiquity of out-of-patent technology in the market place seems to undermine the idea that the public doesn’t benefit from capital allocated to IP R&D. You don’t have to think about it for very long to realize that without these protections, allocating capital to innovate production and synthesis processes that can be maintained a secret becomes the most appealing investment for capital that would otherwise be going to patentable R&D.
A stance against IP protections in general is extremely fringe and extremely short sighted. If a system has problems, advocating for abolishing the entire thing isn’t very sensible, especially when it’s a system that has already provided so much benefit to society.
You might hypothesize that those benefits would have been realized even without that system. But that hypothesis doesn’t really stand up to much scrutiny, especially when we know for a fact that the value this system has created is directly quantifiable.
I think the market has demonstrated that this isn’t what most consumers actually want. But in any case, the anti-piracy provisions of the DMCA are so weak, that anybody who wants to can still pirate with almost absolute impunity. I personally think the anti-circumvention provisions violate the 1st amendment. But it’s hard to say this is a significant regulatory failure, because you can avoid the issue entirely by calling any circumvention tools you produce something other than circumvention tools.
The issues with existing IP laws are readily apparent to me, but this doesn’t extend to an argument against IP as a concept in general. Lots of regulatory systems have shortcomings, but it’s a significant leap to conclude that they can’t be improved.
There are example of things that can’t be effectively regulated, without causing significant harm to society. A lot of people would say that illicit drug consumption falls into this category. But in this case the harm caused by the regulation, and it’s shortcomings in actually providing much benefit is very straightforward. There’s really no comparing a situation like that, and the issues created by existing IP laws. In my anecdotal experience, most of the people who take a broad anti-IP stance also tend to hold anti-private property positions in general.
I think at the point where you agree that IP isn't fulfilling its purpose, the burden is on the concept of IP to demonstrate that it is a valuable legal concept, and all extant implementations simply fail to correctly realize it. Since we're playing the game of making wild extrapolative comparisons to arguments we haven't made, your line of argumentation reminds me of kids who balk at using the Soviet Union as an example for pure command economies being a bad idea.
I think property rights are, if not a necessary part of a functional society, at least something that we don't have a reasonable alternative for. In other places in this thread, I've contrasted situations in which property rights make sense with the ridiculous results of trying to apply the same principles to abstract ideas. The claim that it's not sane or coherent to consider ideas property is in no way equivalent to an objection to property rights in general
I think it very obviously is fulfilling its purpose generally speaking. I think this is evident in that fact that consumers have access to an extraordinary variety of technology, and content. The fact is that the prices for access to this are continually falling, the quality of products and services is continually improving, and these factors are continuously improving quality of life across all sectors of society.
The success of this system, and the benefits it provides to everybody is perfectly evident everywhere around us. You can find instances of this system failing, and identify some patterns in those failures, along with particular components of the regulatory framework that are likely responsible for more bad outcomes than good ones. But the system itself has proved its value many times over.
Edit: I also think the extent to which hoarding content collections has fallen out of favor with consumers is rather compelling evidence that the on-demand model is more popular. But that’s a little besides the point.
Classic
A more reasonable approach would be to look at how resources are allocated today. R&D investments are made in areas where IP protections can be maintained. For technological innovations, this is largely through patents, or by not actually disclosing your IP to the public (by offering your innovation as a service rather than a product to your customers). There are examples of business that have attempted to operate without IP protections, but they've frequently been massive failures that have reverted back to asserting IP protections in order to survive (Mongo, Elastic, Docker...).
There's examples of innovation resourced by non-IP related means, but this is obviously not a very successful model, because the level of innovation it produces is tiny compared to the innovation resourced by IP protection. I can't think of any company that has managed to replicate the RHEL model recently. The give something away for free to drive sales of something else model is quite limited in the ways it can be applied, and often times it's back by selling something with IP rights attached to it. Do you think Intel would put so much effort into contributing to Linux if it didn't have other IP that it wanted to sell you to run your Linux on?
In general, it's very easy, starting with a model and a bunch of phenomena, to explain why all the phenomena fit the model, regardless of what the model is. This is why we care about things like randomized controlled trials or recording when a hypothesis fails to predict a result in science
It's also ridiculous to claim that monetary investment in a context that does have a strong concept of IP demonstrates that it creates more innovation: My whole argument is that the value of IP is to hoard it in order to make money, and companies invest in things that produce it because it will make them money. Given the financial incentives involved, it is vacuously true that this will happen under the current laws, and this has no bearing on whether this is more efficient or produces or encourages innovation
In software, a considerable amount of innovation does happen in FOSS, and gets re-used in industry to build products. This just doesn't get captured as said technologies (libraries, drivers, etc) turning a profit themselves because again, the laws as they currently exist do strongly incentivize, with money, doing things the way the most profitable corporations in this space do it, tautologically
Intel contributing to FOSS is a great example of why a company would care about FOSS even in the absence of IP: Intel sells, first and foremost, a physical good that it has invested considerably in infrastructure to produce, which involves a supply chain, equipment, and material expertise that is expensive and difficult to replicate, much less at the scale that lowers their costs. Without software, that physical product would be of little value to most people, but without patents on the chips, Intel would still be able to manufacture and sell chips and turn a profit
With regard to "Most creative work right now is paid for at production". This is clearly false and almost all creative work is created speculatively, with the hope that it can be made profitable through future sales. All video games, all music, all television, all movies.
The only creative works I know of that are paid for at or before production are things on Kickstarter or those lucky enough to win grants or those with wealthy patrons.
No. The existing definition of "creator" as the person who did the creating is perfectly fine. Various supporters, such as those who provided the creator with lodging, food, salary or other provisions are themselves not the creator.
If you pay an engineer to create a psidget, you haven't created anything. The engineer did. You simply exchanged your money for the engineer's creation. This is similar to how paying a farmer for their crops does not turn you into a farmer.
> not the engineer / tools that were used to create
Do not conflate people and tools.
If you have an idea for a widget, and think there might be a market for it, describe what it should do and what it should look like, then pay an engineer to work out the details and pay them, you are undeniably the creator of the widget.
If Netflix decides they want a show that appeals to 10 year old kids who are into pokemon and video games, then pays people to go work up some concepts, then evaluates the concepts, and green lights the project, then gathers together a team of experts to make it, then there is no question Netflix created the show.
People can be used as tools. It's ridiculous to suggest anything else.
Patents are being granted for trivial things, and the courts are to big and too expensive for individuals to be able to exercise the rights they already have.
All people should have equal power under the law (regardless of wealth), whether its an intellectual property dispute, a malpractice lawsuit, or some development application.
This is a dehumanizing viewpoint.
The idea that creative work is produced "speculatively" rather than "at production" is using your ridiculous prior stated assumption that the entity that pays for the production becomes the "creator" of the work. I've already explained why that's insane. This speculative model of content production is a business model that's emerged in the context of extant laws, not an inevitable necessity of production. It is only possible for very entrenched entities with a lot of starting capital that can afford to take risks in this way, and I don't think it's valuable to protect this business model with criminal law
Or the author of a book.
Its not the capital that gives companies ownership of its creation, its the "will made manifest" by wielding said capital.
The company wants the creation to exist, then the manipulates the world around it so that the thing comes into existence.
You talk of companies as if they are not also people. Companies are collections of people.
The person who had the concept of the invention is the inventor. Not the employer who paid him/her, nor the lab assistants who experimented until they got the concept to work.
For example, in the pharmaceutical world, basic drug discoveries made at universities are transferred to large corporations who typically finance further clinical drug trials, before getting approval and marketing the drug.
Note also, this creates pressure to push the new patented expensive drugs, even if much cheaper out-of-patent drugs work just as well.
Also, a lot of industrial R&D is about improving in-house processes. This kind of research may or may not produce 'intellectual property' but it tends to reduce costs and improve efficiency.
Ideas are cheap, don't let the monopolists corral them all and prevent independent technologies from flourishing.
Pharmaceutical production costs are close to zero. It’s the funding of a research pipeline that mostly fails that concentrates all of the costs. If Pfizer spent 10 years on research and clinical trials of a new drug, and I can the next day of the FDA approval produce a generic derivative of their drug, I will drive them out of business. My costs are essentially 0.
And no, not all innovation has been produced by universities.
In reality we have TONS of examples of competition and rnd coming AFTER something loses IP protection
Everything from Movies to 3D printers can be cited as examples of this, Countless movies are based on Public Domain characters, books, and other works that have entered the public domain, and there is a strong case to be made that the reason studios keep remaking the same movies is because TONS of creative works are locked up in IP hell, there is a HUGE public domain hole due to increasing copy right lengths
For patents, look no further than 3d printers. The tech was there for 20+ years, locked away in various patents. The second those patents expired the industry exploded with innovation and plenty of companies investing all kinds of capital in 3d printers...
https://github.com/minimaxir/hacker-news-undocumented#downvo...
[0]Roughly the equivalent of 'I believe that if <Fiat Currency> ever served a legitimate purpose, it has now run its course and needs to be dismantled. <Because of Bitcoin/crypto>'
[1]Derision intended
I really have to call out Bill Gates who said in 2021 “we can’t just open source the vaccine because factories to make it don’t exist the world over” but he would have been perfectly capable of making those investments in 2020 if what he really wanted was maximum vaccine production. He’s thinking in top down systems and I really think we need to understand how effective bottom up engineering can be, as with 3D printers.
well, the purpose of the patent is to both reward/incentivize and allow the original inventor a monopoly for a period of time to earn back time risk and the development costs, and that monopoly is going to come with higher prices.
it's the point of a patent, you can't very well say that patents undermine the ability to undermine them: it's nonsensical!
I'm not defending patents nor saying we need them, I'm simply saying you can't say "patents don't work because they do what they were designed to"
You must understand, when one inventor gets rights over something, now 1000 other inventors working in a similar field cannot easily build on that idea. So one person benefits while 1000 others are restricted. But the market already rewards innovators, so why stop those 1000 would-be innovators because one person filed all the right paperwork?
And because people always mention this: investment would still work. Instead of a $10m investment for a big five year project, the investments would be smaller and more frequent. Improve the product and impress buyers with the next version. Is it a flop? Someone else will clone your failed version with the necessary improvements, and next year they will get the investment. Product development happens faster because no one can rest on their laurels, and cost of living goes down because no one can charge a very high price for something that can be cloned for cheap.
That faster innovation and lower cost of living is obviously a huge benefit to moving away from patents. This is also a fact rarely acknowledged by those who believe in the standard story about the purpose and benefits of patents. There's billions of people on Earth who need the knowledge we have locked up in patent portfolios. We must not be careless with how we consider this policy. There is a lot at stake.
"To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries."
If the end result of a patent does not promote the progress of science and the useful arts, it undermines the core purpose of patents.
So with both open source and with Shanzai you see that large numbers of people are often motivated to work on the same problem. The function of patents is to prevent those people from pursuing forks of good ideas. This is why 3D printers had gone down in price by 50% after ten years under patent, but in the ten years after the patent expired they went from $25000 to $300 (1.2%). A multitude of curious people will make more engineering progress than one well funded group with a powerful legal monopoly. This is my claim about an under appreciated aspect of patents and this has been discussed in scholarly literature. [2]
A common question I hear relates to funding. People say no one will make investments if someone else can copy their work. But I argue this does not mean investment will cease, but that the nature of investment would change. Broadly it would change from fewer larger investments to more frequent smaller investments. One quarter you get a boost of funding to beat the competition to market, next quarter they copy you but then you copy them back and the cat and mouse game continues. This is actually true capitalist market competition. [3]
So my argument is that patents do not actually accelerate innovation as we are told, but they actually slow down the rate of innovation. And it makes sense: the sole purpose of a patent is to limit innovation for everyone but the patent holder. We were raised on the story that the secondary effects are positive, but that story is not so simple.
Finally, I will say that secrecy can be punished and openness can be enforced. If we all understand how important open source is to rapid innovation, we can make public commitments to boycott those companies and fund efforts to reverse engineer their work. It’s already the case that open source projects avoid proprietary modules as much as possible. We could do this as a society for all goods.
[1] https://www.bunniestudios.com/blog/?p=284
Don't know why you're being downvoted. There are many concrete examples of this.
https://news.ycombinator.com/item?id=28330810
Patent circumvention actively leads to development of new industries in previously ignored locations, leading to prosperity for a huge number of people.
Just compare this to commercial secret practice.
How you can improve that without patents?
https://news.ycombinator.com/item?id=28664769
See also a direct link to a chapter in an economics book about the subject: http://dklevine.com/papers/anew09.pdf
I understand that this is HN and not Congress, but I would also hope that any Government body that is seriously considering this would do a thorough review before passing laws to this effect.
We read articles all the time about patent trolls abusing Texan courthouses to hold real engineers to ransom, and innovation being blocked by patent holders successfully arguing that an obvious innovation was non-obvious (e.g. RED's patent on compressed RAW data), but I don't think many people here would be familiar with the day-to-day, mundane aspects of IP law which probably, for the most part, works fairly well.
The rest of your comment is talking about patents. Was this person's art patented?
"...I don't think many people here would be familiar with the day-to-day, mundane aspects of IP law which probably, for the most part, works fairly well."
I am responding directly to that claim, and have used as examples situations involving patents and situations involving copyright, with the intention of conveying that there is not a single facet of this whole area of law that seems to function well in practice unless your interests are aligned with monopolistic corporate power. Most examples of it not being that bad are examples of it simply not coming up
The fact that there are disputes over who has an IP right means that IP rights are important and valuable and are therefore worth fighting about.
(*) I am a lawyer, but not your lawyer. This is a general observation and not legal advice. Your actual case will be different.
Here's a case I randomly ran across 20 years ago. Sun had recently patented a sequence of three assembly instructions. At the time I read about it, I'd also been reading a 1987 paper that presented that same "invention", and not as a core result, but as an example in an appendix. It had been found by the author's superoptimizer, a kind of brute-force "AI" (to bring it back closer to this thread).
Just to tighten the screws, the lawyer writing the patent put in this passage:
> While embodiments and applications of this invention have been shown and described, it would be apparent to those skilled in the art that many more modifications than mentioned above are possible without departing from the inventive concepts herein. The invention, therefore, is not to be restricted except in the spirit of the appended claims.
So they didn't just appropriate those three lines of old assembly, they made a grab at everything in its neighborhood. And it was granted.
The only part of this that especially surprised me was the coincidence of having that paper fresh in memory just when this patent came up in some discussion online.
(Source: https://darius.livejournal.com/8835.html)
Fundamentally, the operation of "intellectual property" is to treat your intellect as someone else's property under the law. By granting economic monopolies to private companies in particular goods, such as the sale of copies of _1984_, the government converts the abundance provided by our modern machinery into artificial scarcity, thus providing those companies with monopoly profits. The rationale given for this is that it provides an incentive to create those goods in the first place, but where economists have analyzed these incentives quantitatively, it invariably results that the monopoly rights currently granted cause enormously more harm to the public than the extra production they propel.
______
* "Intellectual property" is a fairly incoherent group of laws that it rarely makes sense to discuss collectively. Here I mean copyrights, patents, trade secrets, sui generis database protection, and noncompete agreements, rather than trademarks, mask works, and privacy.
Completely agree. Intellectual property in general needs to be abolished. Copyright in particular is irredeemable at this point.
Stop thinking it's a global problem: just like the regular murder of children at school by other children, it's a very american problem. You can absolutely fix it yourself.
It's particularly ironic on HN; Internet companies are world wide, and the DMCA is used by them frequently. The OP was entirely accurate with his wording, and you're very off base here.
Under DMCA, if someone strikes my classical music composition, I write a one sentence counter notice and that is it. There is no "three strikes" or "YouTube sides with the content producer instead of your counter notice" in DMCA at all.
There are basically only two fixes needed for DMCA: 1) legal liability for false or fraudulent DMCA notices, ideally with punitive damages; 2) platforms that do pretend-DMCA for removing user generated content incur the liability for their actions.
> Under DMCA, if someone strikes my classical music composition, I write a one sentence counter notice and that is it.
Although I agree that there could be problems if a system has no counter-claim process (although, I think that this claim of your is factually incorrect. I think youtube has counterclaims, even for its "fake" DMCAs), I think that you have strongly mis-interpreted the motivations for having "fake" DMCA processes.
The motivations for having fake DMCAs, is that if someone doesn't get a real DMCA, then there is no longer any legal obligation to perma-ban someone for being a "repeat" infringer.
This draconian, repeat infringer policy, which many platforms have, is bad for everyone. And getting around it helps creators.
> "YouTube sides with the content producer instead of your counter notice" in DMCA at all.
I am not sure why you think this is the case. If there is a counter claim, youtube is required by law to respect the counterclaim.
Not if it's the counter-claim in their non-DMCA process.
0: https://www.digitalmusicnews.com/2021/09/15/youtube-music-vi...
What, and stop being able to get free advertising on what is effectively the largest audio streaming website on the internet? In an actually free market, any media company that tried this would immediately get eclipsed by the rest.
But instead, we have a small number of record conglomerates that have formed a trust to bully everyone else to play on their terms, and the US gov refuses to trust-bust until it starts becoming a threat to their political power.
Not just free advertising. A damn lot of these music videos are "monetized", and about of half the ad revenue goes to the studios (if and how much of that money is passed on to the artists... that's another question).
Youtube should do proper DMCA, I fully agree, at least when it comes to the notice-counternotice-lawsuit process defined by the law.
However, it's not as easy. The DMCA is law in the United States, but - surprisingly to some (I have dealt with processing DMCA notices and similar, not for Youtube/Google tho) - it is not a law elsewhere and other nations have their own laws. Even the question of what happens when there is a dispute and party A is in the US and party B is e.g. in Germany is not that easy... Not even what jurisdiction Youtube itself then falls under. Browsing Youtube from Germany gives an imprint that states that the service is provided by Google Ireland Limited (the EU mothership) but also mentions the legal department of Google Germany GmbH (the German subsidiary). Youtube has been sued in German courts over copyright before (e.g. the GEMA case), and Youtube did show up because if they had ignored it instead they would have risked losing the entire German market, if not the EU market.
The kafka-esque system they have right now - with three strikes, ContentID, reporting made extremely easy for alleged copyright owners, challenging these reports made difficult and black box for the channel owners - is most likely a result of their legal department pushing for the global lowest common denominator of any jurisdiction they care about with the smallest risk of Youtube being sued and found liable for user conduct in those jurisdictions.
When the AI can generate a cogent argument for its invention being patentable, then I would be prepared to consider its claim. Until then, comparisons to the DMCA seem overblown.
Scenario: Microsoft owns copyright over AI and makes AI available as part of windows. Are all "inventions" made using windows now also partially created/owned by Microsoft? Nobody would say that Microsoft should get rights to a book because the author using Word's spellchecker, but what about an AI-based translator program? Is the AI a co-author? If so, Microsoft could argue that it is a part owner of everything created using an AI-enabled MSword.
Scenario: Photographer takes picture of mountain. They get copyright because the contributed to the setting up of the camera. They pick the scene and they pushed the button. But new AI camera, with AI trained on all the best photographs, is now in control. Autofocus does not make the camera a creator, but an AI that has a hand in every detail of the photograph? An AI that is legally allowed to be a copyright owner? Is every photograph now a joint work between human and AI? There are lots of iPhone cameras out there. What is Apple's stance on AI creators?
I suppose that depends on what crafty Microsoft lawyers have written to the contracts. I would assume no. It's a tool which is in the possession of the person who bought the license from Microsoft.
I bet a billion people getting word bundled with O365 would still use it, mostly because no one would have read page 3456 of the ToS and wouldn't know any better.
Some nerd packs online would scream it from the rooftops but 98% of the population wouldn't care.
Startups are already in this boat of growing percentage of value created allocated to the platforms of distribution or systems that the ideas runs on.
Can an AI sign a patent assignment document or a work-for-hire contract? What would it mean for an AI to sign such a document "freely"?
And, Michaelangelo used some tech (scaffolding) while painting the ceiling of that chapel (Sistine) where the catholic bishops hang out while choosing a new bishop of Rome. Reliable scaffolding was certainly enabling tech for that project. It doesn't mean the scaffolding gets a credit.
Patents though are a disaster for humanity and are used to create corrupt monopolies that overcharge by many billions of dollars because they have no competition. Further patents stop new innovation because companies like Boeing own all the patents related to flying commercially viable airplanes. You might have a good idea for a new airplane company, but you can't do it because Boeing owns all the patents on everything that you will need to build a flying plane.
That said, IDK if AI-invented patents really pose a challenge. It might be a philosophical challenge to the idea of patents, but that doesn't necessarily translate into a challenge in practice.
IMO logically, if machines cannot be inventors than anything invented by machine is arguably trivial. However, I don't think this kind of reasoning translates to a legible legal argument.
A mistake, or a resounding success in achieving exactly what you mention? Do you really think the DMCA was passed with the people's interests in mind?
Using a mechanical process to search a large possibility space to find a patentable subspace is not new. For example, testing chemical compounds to find a mixture with optimal properties or testing drug dosages to find the most effective treatment is common. The resulting narrow range of mixtures or dosages is patentable so long as it is new (not done before), non-obvious (inventive), useful, etc. Also, if I hire other people to do the lab work at my direction, then I am still the inventor, not the lab technicians. Similarly, AI is just another tool.
Edit: (*) After looking into this more, the applicant here is on a crusade to have an AI recognized as an inventor. https://artificialinventor.com/patent-applications/. That web site contains nonsense like the following:
> Arguably, DABUS may be considered “sentient” in that any chain-based concept launches a series of memories (i.e., affect chains) that sometimes terminate in critical recollections, thereby launching a tide of artificial molecules. It is these associated memory sequences, and the accompanying simulated neurotransmitter rush, that are considered equivalent to subjective feelings in humans (i.e., sentience). In this way, DABUS has an emotional appreciation for what it conceives.
"Family name - GPT, ethnicity - X86/ARM/NVidia?, Birth date - [not an adult?], gender... fuck that. Joe, tell them no, I'm not doing this shit!"
It’s all nonsensical anyway.
Wait what? Source?
In terms of evidence: the simplest argument is probably that humans appear to consistently obey the laws of physics, and the laws of physics appear to be mathematical.
Until we have this technology should we adjust laws for hypotheticals? No.
We don’t have a complete theory of physics. We don’t have any viable models of human thought. This is a non sequitur.
You’ve reduced the question of whether our universe is deterministic, something the greatest minds in physics have debated for a century and continue to debate, to a sound bite. That’s disingenuous. There may be phenomena we cannot accurately model and predict for fundamental reasons that have nothing to do with the supernatural.
We are so far from scientifically understanding consciousness that we don’t even have an agreed-upon empirical definition for it, let alone a way to measure it. That gap in understanding undermines any claim that it can be modelled this way or that.
Consciousness is more interesting, and it boils down to the claim that there is something metaphysically and unverifiably (from the outside) special about humans. That sounds supernatural to me.
If this is the case why are you arguing on HN? Is it possible for you to leave the discussion? Is talking to you the same as dropping a ball and watching gravity pull it downwards?
I'm a little skeptical of your stochastic model of humanity as well. Statistical mechanics works because oxygen molecules don't go off and start religions and convince other oxygen molecules to behave radically differently for thousands of years (or at least we haven't ever seen it). In fact our stoachstic physical models, classical, semi-classical and quantum, kinda specifically preclude this. How would you design an equation to capture this real phenomenon in humans?
No you cannot, quantum effects are way more complex than just random chance effects. The quantum waves are particles in their own rights that flows and interacts with each other and itself, if you treat it as just a stochastic variable you don't get quantum effects.
Even with indestructible information, combined with locality at the big-bang, combined with a perfect fundamental physics, is there enough time/energy to compute that? Perhaps this is grasping at straws.
With regards to the law, you’re saying, because we assume that one day we should have a complete simulation of humans, or at least a model, therefore, we should update our laws NOW, to prepare for this?
A story of premature optimisation if I ever heard one.
It is possible you can make a good simulation of human thought without all of that, but we can't be certain.
> At first it may appear unlikely that a complex system with many degrees of freedom like the brain could be modelled with the right causal dynamics, but without taking into account the smallest parts … However, state variables of complex systems can be quantitatively predicted when there is ‘scale separation’: when different aspects of the system exist on sufficiently (orders of magnitude) different scales (of size, energy, time etc), they can become uncoupled … [an] example is the scale separation between electric currents and logic operations in a computer …
> When there is no scale separation (such as in fluid turbulence) macroscale predictions become impossible without simulating the entire microscale. An important issue to be determined is whether such a cut‐off exists in the case of the human brain … if it could be demonstrated that there is no such scale, it would demonstrate the infeasibility of whole brain emulation.
The FHI is assuming scale invariance in order to test it experimentally. We shouldn’t assume it for the purpose of analysing the roles played by humans and AI under IP legislation.
[1]: https://www.fhi.ox.ac.uk/brain-emulation-roadmap-report.pdf
It's a very common worldview among the tech set.
Edit: Uh, downvotes, really? I think it's totally reasonable to assume this as a given since it's, like, pretty obviously true. If anything the parent needs to provide their own extraordinary evidence for why consciousness would be non-computable. I'm just trying to provide some context for them to read up more on some stuff that apparently everyone but them (who has given it any thought) already knows about.
The algorithm we run on is special, but it's physically encoded and self generating, based in our DNA and physical development. We're vastly complex but entirely built of relatively simple fundamental building blocks.
We are computers, in the abstract sense of Turing machines and such concepts.
[citation needed]
I am not an expert, my understanding of neurons is largely is based on [0], [1], [2].
[0]: https://mitpress.mit.edu/books/principles-neural-design
[1]: http://www.sci.utah.edu/~macleod/bioen/be6003/notes/W08-spru...
[2]: https://mitpress.mit.edu/books/functional-connections-cortic...
Where an impulse comes from, when it is sent, the current charge state of the receiving neuron, the specific neurotransmitters released by the axons, and the efficiency of signal transmission at a given point in time all play a part in the signal sent along dendritic, but at any given point in time a neuron is firing or isn't, and that's the only state that matters to the other neurons it's connected to.
The biggest difference between biological brains and artificial neural networks is the analog nature of biological timing compared to the arbitrary cycles of software.
Spike timing dependent plasticity networks address some of the problem of variable timing and its impact on signaling.
https://en.wikipedia.org/wiki/Spike-timing-dependent_plastic...
At any given moment, you could capture the state of a biological brain as a binary number. The model would need to account for all sorts of analog mechanisms and a huge, biologically messy graph of connections with dozens of variations across a handful of neuron types, and the length of a synapse and where it physically connects or touches other synapses without connecting directly and... lots of stuff that's hard to model and even hard to see.
I'm not arguing that every part is digital or even equivalent to computers on silicon, simply that the basic mechanism of computation is itself digital pattern processing. It literally cannot be anything else.
There is no evidence or theory to indicate any other plausible method of operation. All the analog stuff can be swapped out with an arbitrary bit length representation and the functionality is retained. Remove the digital assumption and everything breaks. You need magic or spirit to make it work, which is ridiculous because silicon computers proved that binary computation is a perfectly fine way to represent and process data. Quantum processing in biological brains is magical thinking performed in a desperate attempt to ascribe eminence to human cognition, because for some reason people don't like the idea that brains do digital processing.
We are messy and vastly complex and sophisticated digital pattern processors. Which makes evolutionary sense, because that's how our predecessors work, all the way back to single cell organisms and self replicating DNA chemistry.
Signals and patterns in the real world can be represented as binary to an arbitrary precision. The Planck length dictates the maximum level of precision relevant to our universe.
https://www.researchgate.net/post/How-many-bits-of-informati...
Human brains achieved a special state of depth and breadth of processing, concurrent with physically encoded, evolved algorithms in our neural architecture. The arrival of biology at human cognition affords us the privilege of deep reflection, the apparent subjective reality we experience, and the thus far unlimited scope of conception such that we can make approximations about the size of the universe or the mass of a photon. Or the operations of the minds we possess.
We operate in binary - the explanation is sufficient, simple, and corresponds to all known evidence. There aren't any outstanding questions requiring that premise to be questioned.
It's a superficial and unfounded observation that tries to frame neural computation as a bias or conceit of modern culture. Not how science works, and the argument is condescending as hell.
But I agree that it's a really shallow way to dismiss arguments. In this case, it's basically just saying that a brain is deterministic and made out of parts. Clockwork and steam engines and circuits all work fine. I can't comment on the catapult without more elaboration.
Without getting into the weeds too much, I just find the certainty of the systems line of thinking a little unfounded.
Consciousness is an algorithm performed by a computer with messy biological parts, some of which are analog, which may have serious implications for the level of strictly digital computer power you need to successfully implement the algorithm in silicon. The same applies to intelligence, and it's an open question as to what exactly the relationship between subjective experience and intelligence is.
You are a brain in a bone vat experiencing a subjective construct produced by a biological computer processing binary states consisting of internal feedback and sensory streams made entirely of patterns distinguished at any moment in time by the presence or absence of electrical impulses. Each neuron can be receiving a pattern of up to ten thousand individual bits, and whether the neuron fires or not depends on the current state and precise pattern of inputs.
Everything you experience, think, or do, is determined by those electrical patterns.
However, the universe doesn’t run on math, it runs on itself. Math is the best way we have to describe it at the moment. It used to be that divination was the best method. Why are you so certain math in it’s recognizable form today will be the method used to describe the universe in 100k years?
Our brains certainly do not run on binary. I think you might be confusing the information content, represented in bits, with the actual implementation. At the very least dendrites take in and mix signals of various strengths. The threshold for triggering and strength of the action potential are dependent on many chemicals signals. And that’s even a dated understanding which gave rise to the ideas of artificial neural networks. I’m sure the current understanding is much more complex and nuanced.
You brush off the analog parts as an implementation detail, when it could have very real implications for the computability of consciousness. The thing is, we don’t know.
But even all that is besides the point. The main point is, we don’t have the slightest idea how consciousness works. To the point that no one can even figure out how to ask the right questions about it.
In the face of such great uncertainty, it is anti-scientific to state with such conviction that you have the correct model. It’s fine to present it as a hypothesis, but the certainty is unfounded.
https://www.julianjaynes.org/resources/books/ooc/en/introduc...
The way he does it is quite compelling - we do make metaphors of the mind to current tech. This was written well before current computer neural networks, and gives a good account of the sound version of this argument.
From now on, nothing is my fault any more.
I can reduce all the music in the history of the world to the phrase "rhythm, melodies and lyrics". So? That'd say nothing about all that music, it'd just say something unflattering about me.
> It is not the greatest of modern scientists who feel most sure that the object, stripped of its qualitative properties and reduced to mere quantity, is wholly real. Little scientists, and little unscientific followers of science, may think so. The great minds know very well that the object, so treated, is an artificial abstraction, that something of its reality has been lost.
-- C.S. Lewis
There is no scientific evidence whatsoever for this very bold claim.
The only escape hatch is to assert that humans can't be described by physics.
Or that there is some physics that can’t be described by computation. It seems vanishingly unlikely, but there is no actual proof that everything physical is computable. And even if there was, it probably has nothing do do with minds.
That might seem like a fine distinction, but the importance going forward is stark given a quote from TFA:
> The third judge, Lord Justice Birss, took a different view. While he agreed that "machines are not persons" ...
Would you still say that if said "conscious beings" were in a desperate struggle with humanity for scarce resources required to survive?
Progress has boundaries and hits walls. Such a breakthrough in AI is not guaranteed to happen.
If we create a box of metal, matrices, and silicone and tell ourselves it’s conscious it will be our own hubris that leads to additional competition for shared resources and further human inequality/suffering. Such lines of reasoning make me very concerned.
First, different researchers or their funders presumably have very different motivations. Compete with China, fire all our employees, become famous as an inventor, make lots of money, work on something interesting ...
Second, at some point AI's are likely to have their own goals and "help humans" quite famously is not guaranteed to be their North Star.
Only Humans, Not AI Machines, Can Get a U.S. Patent, Judge Rules - https://news.ycombinator.com/item?id=28405333 - Sept 2021 (7 comments)
South Africa issues world’s first patent listing AI as inventor - https://news.ycombinator.com/item?id=27995313 - July 2021 (75 comments)
EPO and UKIPO Refuse AI-Invented Patent Applications - https://news.ycombinator.com/item?id=21990346 - Jan 2020 (39 comments)
While I agree with the current ruling in the UK, this statement does not sit too well with me:
> "Only a person can have rights. A machine cannot," wrote Lady Justice Elisabeth Laing in her judgement.
In my opinion this sets a bad precedent in case we ever achieve artificial general intelligence (AGI) [2], which I think is perfectly possible, especially considering that we humans are nothing but complicated biological machines. And I think an AGI should very much be considered a person. That's why I think the way how a US judge in a prior cases put it is more agreeable:
> As technology evolves, there may come a time when artificial intelligence reaches a level of sophistication such that it might satisfy accepted meanings of inventorship.
> But that time has not yet arrived, and, if it does, it will be up to Congress to decide how, if at all, it wants to expand the scope of patent law.
But admittedly this is still all very hypothetical as I don't see AGI happening in the near future and for now there is no real problem.
[1]: https://en.wikipedia.org/wiki/The_Measure_of_a_Man_%28Star_T...
[2]: https://en.wikipedia.org/wiki/Artificial_general_intelligenc...
The fight will then be to get AGI classified as a person. If a corporation can have personhood, it is not impossible for AGI to have the same.
So you have inventor: Steve Wozniak; assignee: Apple Inc.
Assignment allows the owner to be different from the inventor.
The basis for corporate personhood is that it is made up of humans. The benefits of a corporation flow to real humans. Real humans actually direct how a corporation will be run. In the event of crimes, the corporate veil can be pierced to go after the real humans behind it.
The “personhood” of corporations flows from the “personhood” of the humans behind it.
Imagine a world where corporations held no rights at all. For example, while an individual human had the right to freedom of the press, a corporation like the New York Times would not have that right.
Anyhow:
It's impossible to talk about patent reform without first separating software from everything else. Otherwise, you're trying to make standards that apply equally to bankruptcy and stand-your-ground laws. They're just different.
If Congress and/or the courts made software not patentable once and for all, then the patent system would look completely different. Most HN readers would stop thinking about patents altogether and move on to copyright.
When “AI” says no I don’t want to invent a patent I want to play music (or whatever else it may desire) I’ll believe it exists.
Machine learning, deep learning, neural network, and gradient decent are better terms but I’m not an expert.
Even companies cannot be assigned as inventors, i.e. „Apple“ has 0 inventions. They can only be assigned the commercial rights to what a human invented.
The correct way of course would be to lobby for a change in the law, via democratic means, if you really think an AI can invent something.
Why did anyone, anywhere, at any point in time, believe that it was reasonable to criminalize programmers for writing the correct solution to a problem?
Every single lawyer that has ever defended software patents should be immediately disbarred, and forbidden from ever practicing law again.
In other countries the inventor can be a corporation. There are some cases where this matters (e.g., disputes over inventorship or ownership of the invention of an employee), but in most cases it probably doesn't matter.
He is filling similar cases all around the the globe and is (mostly) loosing.
It remains unclear whether his motivation is financial or ideological or something different all together.
If you invent something and someone else patents it, you show prior art to invalidate the patent.
Whether you used software or a lathe or a lawnmower or excel to make the thing doesn’t matter.
(Not a lawyer)
There's an exception: When a second applicant disclosed the invention publicly both prior to the first application and less than one year prior to the second application.
Another note: Prior art needs to be publicly available. So, a first inventor's private notes can't invalidate a second inventor's patent. In the old first-to-invent system, however, the private notes of a first inventor could allow the first inventor to get the patent ahead of a second inventor who filed first.
If you were running it and someone stole the results prior art would likely be yours. If you allowed someone to run it with their own inputs it would likely be theirs.
It's probably up to the bank, and of course rather challenging for them to pay their income taxes as we all must ...
¹: for the sake of argument, assume this
Cryptocurrency is basically designed for this.
most definitely, yes. why wouldn't it be the owner?
The person who pushes the shutter button is the ‘inventor’ of the picture.
Sadly, AI will object this in court and win on all counts.
Thaler recently won the same argument in Australia [2], although an appeal is pending. His lapsed patent application (2019363177) can be found in IP Australia’s online database [3]. Putting the legal question aside – whether Thaler’s insistence on naming DABUS the inventor was fatally defective to the application – the patent application is pseudoscientific nonsense, not a sound basis for deep questions about the nature of ownership and invention.
The patent claims two apparently unrelated inventions. It does not explain how they were identified by DABUS or identify any other connection between them. The first invention is a “food or beverage container” with the cross-section of a Koch snowflake, which supposedly allows for containers to stick to each other. The material used for this flexible, fractal surface is unspecified, but plastic, metal or “an elastomeric material” are suggested, along with “potentially flexible food product (for example pasta, dough, liquorice and so on.” It is plausible that the text of this claim was generated by GPT-3.
The second invention is a “neural-flame device for attracting enhanced attention,” that is, an LED that blinks “at a frequency corresponding to approximately 4 Hz and a fractal dimension near ½.” There is extensive citation of Thaler’s own publications, none of which are in reputable peer-reviewed journals. What can be found online [4] fails to show that flickering LEDs have some useful neurological function, or explain what the patent application has to do with DABUS.
[1]: https://www.bailii.org/ew/cases/EWCA/Civ/2021/1374.pdf
[2]: https://www.judgments.fedcourt.gov.au/judgments/Judgments/fc...
[3]: http://pericles.ipaustralia.gov.au/ols/auspat/applicationDet...
[4]: https://www.sciencedirect.com/science/article/pii/S187705091...
but on a serious note, why do people say things like "humans are nothing but complicated biological machines"?
i mean, why do you really want to replicate the human brain?
i don't get it. is it alienation, loneliness? projecting your hope for connection and understanding onto something you program? is it to stay in control and avoid disappointment?
today too many vital feedback loops of science and technology are withheld from the commons by the propertied class. imagine what we could do if all science was mapped, further developed and universally shared in the open; no longer monopolized behind corporate walls. that's something that excites me a lot. not black box 'AI' owned by Musk, Zuck, Brin, Page, Cook or whomever. seriously, science fiction about this whole AI thing is unrealistic and dystopian.
when we start to realize most science and technology was publicly funded, will we seriously consider ditching the criminalization of knowledge sharing (intellectual property laws)?
“Many of the problems being faced today by the [US] are due to the fact that US taxpayer...do not realize that corporations are making money from innovation that has been supported by their taxes.” That they are not aware of the benefits to competition seems to be a triumph of free-market ideology over good sense.
How many Americans are aware that Google’s basic algorithm was developed with a National Science Foundation grant?" [1]
do we really want to destroy the world so we can have a 'Free Guy' Ryan Reynolds -like fake AI friend?
"90% of machine learning is burning down a forest so that you can train a computer to do phrenology" [2]
— @cooljqln
[1] https://web.archive.org/web/20160204223931/https://nybooks.c...