Month: October 2024

If AI Tramples Copyright During its Training and Development, Should AI’s Output Benefit from Copyright Protection? Part One: Stephen Thaler

” A Recent Entrance to Paradise”, Stephen Thaler (not protected by copyright)

One of the ongoing debates about works made with generative AI is whether they qualify for copyright protection. Should they? Let’s consider the essence of copyright. What is its raison d’être? According to the classical European definition, it is to respect the property rights of the author (droit d’auteur), sometimes described in the simple terms of the Eighth Commandment (“Thou shalt not steal”). According to the more utilitarian Anglo-Saxon rationale for copyright, it is to benefit society by rewarding and incentivizing authors, thus stimulating further production of works for the greater good. In either case, it gives one pause to wonder how works created by AI fit into either school of thought. Is there an inherent property right in content produced by an algorithm? How does copyright protection incentivize an algorithm to produce more “useful arts”?

In my view, the only way one can square this circle is by attributing human creation, or at least a degree of human creation, to AI generated works. But this opens Pandora’s box; if there is to be human attribution, to whom in the chain of creation does the credit fall? How much human creativity is required? It also raises the spectre of hypocrisy, as the AI industry hijacks the creative work of others, without recognition or recompense, yet has the gall to claim that AI outputs are unique and worthy of protection. I have written about these issues before (here, here, and here), but a couple of recent cases in the US have brought these fundamental issues bubbling back to the surface.

When it comes to trying to prove that AI and copyright protection go together, there are a couple of different approaches. One, most notably espoused by an AI technologist in the US named Stephen Thaler, is to claim that a given work was produced exclusively by AI but should nonetheless be protected. In Thaler’s case, the work for which he is seeking copyright registration was created by a particular AI “machine” (or algorithm), specifically the one he “invented”, the so-called “Creativity Machine”. Thaler claims his machine should hold the copyright but behind the machine, of course, stands Thaler. This is not dissimilar to existing British copyright law where, under Section 178 of the Copyright, Designs and Patents Act, 1988, works “generated by computer in circumstances such that there is no human author of the work” are nevertheless accorded copyright protection for fifty years from date of creation, with the copyright being held by “the person by whom the arrangements necessary for the creation of the work are undertaken”, even if there was no creative act undertaken by that person.

As I noted in an earlier blog post (The Humanity of Copyright), Thaler began his (so far) unsuccessful pursuit of US copyright registration for his professed 100% AI generated art work, “A Recent Entrance to Paradise”, back in 2018. Despite several reversals, both in the application process at the USCO, its Review Board and in the District of Columbia courts, Thaler persists in his quixotic journey. He has unsuccessfully argued various precedents for non-human copyright ownership, such as the “work for hire” doctrine and corporate copyright, although it is worth noting that humans stand behind both. He is now apparently pursuing the common law theory of “fruit of the tree” in his attempt to get the USCO to register his work. To my non-legal mind, this is the ultimate stretch.

What Thaler could do is to claim that at least part of the work is the result of his personal creative efforts. That was the USCO outcome for the graphic novel Zarya of the Dawn produced by writer Kristina Kashtanova (who identifies as “they”). The novel contained both generative artwork and human story and design elements. After initially registering Kashtanova’s work, the Copyright Office cancelled the registration after they (Kashtanova, that is) claimed it was AI produced. The Office subsequently reconsidered and granted copyright protection to the parts of the work Kashtanova had created, namely the text, and selection and arrangement of the work’s written and visual elements. That, however, is a step too far for Thaler who continues to push for recognition by the Copyright Office that a work produced exclusively by his “Creativity Machine” can be protected by copyright. That seems very unlikely to happen.

While Thaler doggedly pursues copyright registration for “A Recent Entry to Paradise” (featured on this blog post—after all, it is not copyright protected), others who have created art using AI are following a different track. One of these is Jason B. Allen, whose award-winning digital art creation “Théâtre D’Opéra Spatial” has also been denied copyright registration. Allen’s approach is the opposite to that taken by Thaler. In contrast to Thaler’s insistence that the work is a creation of AI (his AI “machine”), Allen insists that he is the source of the creative inspiration behind the work, notwithstanding that it was created by an AI algorithm, Midjourney. Allen’s pursuit of copyright registration for “Théâtre” will be the subject of my blog post next week.

© Hugh Stephens, 2024. All Rights Reserved.

Singapore Inhibits Rightsholders Ability to Use Contracts to Prevent Unlicensed Text and Data Mining of Content

Image: Shutterstock (AI modified)

Singapore already has one of the most permissive text and data mining (TDM) exceptions in copyright law found anywhere, allowing AI developers to ingest copyrighted content for AI training purposes subject only to a very few limitations, all of which are pretty minimal. These provisions were introduced back in 2021 when several changes were made to the Copyright Act, some, frankly, better than others from the perspective of rightsholders. The TDM exception is not among the more positive outcomes. It is made worse by a provision in Singapore law that, for all practical purposes, prevents rightsholders from exercising contract terms to prevent the unlicensed appropriation of their content for the commercial benefit of third parties, namely, AI developers.

The term used to describe text and data mining in Singaporean law is “computational data analysis”.  As explained in this law firm blog post, this is defined as including;

“using a computer program to identify, extract and analyse information or data from the work; (or) using the work as an example of a type of information or data to improve the functioning of a computer program in relation to that type of information or data – a specific example being the use of images to train a computer program to recognise images”

The exception also permits supplying the works to other persons, provided this is for the purpose of “(i) verifying the results of the computational data analysis carried out by the latter; or (ii) collaborative research or study relating to the purpose of such analysis carried out by the latter.”

This is simply commercial, unlicensed web scraping of copyrighted content for AI development by another name, with very few limitations.

Singapore is now contemplating opening that door even wider by permitting the circumvention of digital locks (aka Technological Protection Measures, or TPMs) to allow text and data mining for AI training, as I wrote about here. (Singapore’s New Copyright Act Three Years On: There’s No Need to Open the AI Exception Door Even Wider). This proposal needs serious reconsideration as it would seriously tilt the copyright balance in favour of AI platforms to the detriment of rightsholders.

Another example of Singapore’s increasingly permissive approach to copyright protection is its undermining of the sanctity of contracts by limiting contractual terms that prevent unlicensed and unauthorized use of copyrighted content. In other words, the Act limits the ability of contractual terms to protect against, or override, the TDM exceptions. Rightsholders cannot “contract out” of the exceptions.

This provision was included when the Copyright Act was last updated (2021). The list of exceptions that cannot be restricted by contract was expanded to include the exception for computational data analysis use, as well as a couple of other uses; use of a work in judicial proceedings or for legal advice, and functions of galleries, libraries, archives, and museums, the so-called GLAM sector.  The computational data analysis exception is the one of concern. It requires that to be enforceable, a contract that limits the ability of a user to scrape content must be individually negotiated. In other words, standard contract terms on websites that limit use (so called shrink-wrap or clickwrap agreements) cannot be used to override the TDM exception. This has the effect of rendering standard contractual terms virtually unenforceable. They become the exception rather than the norm.

The term “shrink-wrap agreement” was originally applied to the preprinted agreement included as part of the packaging containing a software program. By opening the packaging, the user agreed to comply with the licence terms of the software. The term has since been expanded to include “clickwrap agreements” that take effect when a user accepts the terms and conditions of a website. This can be used to specify that the content about to be accessed can only be used for certain purposes or under certain conditions. One of these conditions could be a restriction on use of unlicensed content for AI development. The Singapore legislation eliminates what is a standard practice used by rights-holders in many parts of the world to protect and control use of their content. It also means that robot.txt files used by rightsholders to signal that their content should not be freely scraped (compliance is voluntary) are unlikely to be respected in Singapore. Robot.txt limitations are often included in clickwrap agreements.

Not only does the Singapore law allow for a broad undermining of contractual terms, and prevents “contracting out”, but its TDM exception is very wide in terms of application. In the UK, while contractual terms cannot override the TDM exception, unlike in Singapore allowable TDM use is much narrower. The exception in the UK can be used only “for the sole purpose of research for a non-commercial purpose”. No such restriction exists in Singapore. In the EU, contractual terms can override the general TDM exception (Article 4), unless the unlicensed access is “conducted by research organisations and cultural heritage organisations” or is “for the purposes of scientific research”, (Article 3). In these limited cases only, the contractual override does not apply. This still provides broader protection for rights-holders, and where the contractual override is disallowed, it is for very limited purposes. This is a much more nuanced approach than the one adopted by Singapore.

Contract law is generally seen as the oil that lubricates the wheels of business. In the digital age, shortcuts in the form of clickwrap agreements have been used to convey contractual terms to users. In some jurisdictions, explicit consent is required by clicking “I Agree”. Singapore’s current copyright legislation undermines the sanctity of contracts by imposing unrealistic conditions, particularly with respect to limiting the rights of rightsholders to prevent web-crawlers from ingesting copyrighted content without licence or permission. To say this is problematic is an understatement.

Singapore can do better. As an exemplar of rule of law in the region, it should be as assiduous in protecting the rights of copyright owners as it seems to be in advancing the interests of AI developers. The motivation, apparently, is to promote “innovation”. This is a misread of what brings about innovation. True innovation comes with a partnership between rights-holders and users that protects and compensates rights-holders for the time, effort and investment they have put into developing content that is clearly of value to the AI community. That content should be licensed, or at the very least, rights-holders should be given the option to opt-out through the ability to enforce contract terms, including overriding text and data mining exceptions when necessary.

© Hugh Stephens, 2024. All Rights Reserved.

Smoking in Films: Governments Should Do What is Reasonable and Responsible (Indonesia Please Take Note)

Image: Shutterstock

I am a lifelong non-smoker, although this is more by luck than design. When the other kids were lighting up behind the school, I joined them. However, while some of them got over the feeling of nausea brought on by those first puffs and became hooked on nicotine for life (hopefully, to their “undying” regret), I was never able to get over the initial hurdle. I was lucky and claim no moral superiority. Today we all know about the public health consequences of smoking. Governments have developed a variety of policies designed to discourage smoking, from banning indoor smoking, to high taxes, to plain packaging, to health warnings and other educational efforts. This has had a high degree of success.

Yet despite a couple of generations of anti-smoking campaigns, smoking is still a reality in our society. Peoples’ motivations are different, and it is not for me to judge or preach. They are fully aware of the risks. Banning smoking would be about as effective as Prohibition was in reducing alcohol consumption. It had just the opposite effect. For this reason, proposals to make the sale and consumption of tobacco products illegal have understandably gone nowhere. The same is true for regulations that pretend smoking doesn’t exist. For better or worse, it does, and is a part of everyday life, today and in the past. And this brings me to the point of this little trip down the highways and byways of anti-smoking efforts; smoking and its relationship to filmed entertainment.

One cannot think of Humphrey Bogart without a fag hanging from his lips, a trail of smoke wafting across the face of his paramour of the moment. It would be impossible to reflect the atmosphere of a smoke-filled saloon in Dodge City without the presence of roll-your-owns. And what would W.C. Fields be without a stogie? However, it is not just historical films where smoking plays a part in creating the scene and depicting reality. Whether we like it or not, smoking is a part of the current scene (especially, it seems, in Europe). If any readers have sat recently in an outdoor café in Spain, France or Italy, you will know what I mean. I am not saying this is right; I am saying it is a fact of life.

This raises the question of how the film and TV industry should deal with the smoking issue. Should it pretend it doesn’t exist or should it deal with the issue responsibly? For decades the industry has grappled with this dilemma. The current approach is to eliminate smoking in children’s programming and issue clear warnings with respect to content aimed at more mature audiences while depicting smoking on occasions where it is necessary to convey the story. The filmed entertainment industry builds its product by telling stories that, for the most part, reflect the real world in one way or the other. Yes, there are films set on the planet of the apes, or in some future world, but most stories take place in a contemporary or historical setting. Being a reflection of society, at times filmed content depicts people smoking.

Depiction, however, has nothing to do with promotion. They are totally different.  Yet despite the difference, there are proposals from time to time in some countries–such as Indonesia at present–to deal with smoking as if it doesn’t exist. The draft regulations as currently framed would ban all scenes in filmed content that depict any smoking. This includes content produced by international streamers, such as Netflix, Amazon Prime, HBO, Disney +, etc. This, I would submit, is a step too far—and completely unnecessary.

The draft regulation is in preparation for Indonesia’s accession to the Framework Convention on Tobacco Control under the umbrella of the World Health Organization (WHO). Many countries in the Asia-Pacific region, and elsewhere, are already signatories. The treaty has been in force for about twenty years and is designed to coordinate international tobacco control efforts. According to the WHO, it has been an effective instrument in curbing what it calls a “global tobacco epidemic”. It is a worthwhile endeavour, worthy of support. The treaty was initially signed by 168 countries and now has 183 Parties, representing 90% of the world’s population. Indonesia is one of the few not to have yet acceded and the only one in Asia.

Accession and ratification requires compliance with the obligations of the Convention. These include restrictions on tobacco lobbying, demand reduction (taxes), protection from passive smoking, product regulation, packaging and labelling (health warnings), public awareness, advertising bans, addiction cessation programs, actions against smuggling and restricting sales to minors. These measures have all proven to be highly effective. Notably, there is no mention of depiction of smoking in films or TV programming nor any requirement to censor it.

However, Article 24 of the draft Indonesian Health Regulations regarding tobacco and electronic cigarette product safeguards prohibits the broadcasting, depiction, show or display of individuals “smoking, showing cigarettes, cigarette smoke, cigarette packaging, or any product relating to Tobacco and Electronic cigarettes in print media, broadcast media or information technology media related to commercial activities, advertisement or encouraging smoking.”

The draft regulation then goes on to say;

“The prohibition…includes streaming services that offer TV shows, movies, anime, documentaries, and other content accessible via internet-connected devices…Streaming services include Netflix, Disney+ Hotstar, WeTV, Viu, Iflix, Prime Video, GoPlay, CATCHPLAY+, MolaTV, Twitch, and others.”

Let’s remember that international (and domestic) streaming services are premium offerings directed at a willing, paying audience. A viewer has to “opt in” by paying a subscription. Moreover, content on the streaming platforms already complies with the anti-smoking limitations described above, such as warning labels and no exposure to children. If implemented, the Indonesian decree would require international streamers and their domestic distributors to edit all content to remove any and all depiction of smoking. Not only would this emasculate content that viewers have paid to watch, it is both impractical and unnecessary.

The draft regulation is not an accidental sideswiping of the streaming industry. It is aimed directly at it. It is hard to understand why. It already deals responsibly with smoking on screen. Implementing this measure would simply shift audiences to unedited pirate content, which is rampant in Indonesia.  

The relevant section of the WHO Convention is Article 13, covering tobacco advertising, promotion and sponsorship. That article, along with its Interpretive Guidelines, requires acceding states to control advertising for tobacco products by prohibiting advertising that is false, misleading, or deceptive and by requiring health warnings on tobacco advertising, promotion and sponsorship. It also requires restrictions on incentives to purchase tobacco and prohibition, if legally possible, of tobacco sponsorship of international events and activities. Again, nothing about banning depictions of smoking.

The 28-page Guideline document goes into more detail. It raises the issue of tobacco in media, noting that the tobacco companies can get very creative when it comes to promoting their products, sometimes using “hidden forms of advertising or promotion, such as insertion of tobacco products or tobacco use in various media contents”. In other words, product placement. However, depiction of incidental smoking in films has nothing, absolutely nothing, to do with product placement or tobacco promotion.

The Guidelines go on to list what the elements of a comprehensive ban on tobacco advertising, sponsorship and promotion, as required by the Treaty, should cover;

■ all advertising and promotion, as well as sponsorship, without exemption;

■ direct and indirect advertising, promotion and sponsorship;

■ acts that aim at promotion and acts that have or are likely to have a promotional effect;

■ promotion of tobacco products and the use of tobacco;

■ commercial communications and commercial recommendations and actions;

■ contribution of any kind to any event, activity or individual;

■ advertising and promotion of tobacco brand names and all corporate promotion; and

■ traditional media (print, television and radio) and all media platforms, including Internet, mobile telephones and other new technologies as well as films

In other words, both traditional and non-traditional media should be subject to the ban on tobacco advertising and promotion. But this has nothing to do with incidental depiction.

Drilling down a bit further, there is a section dealing with depictions of tobacco in entertainment media, followed by Recommendations;

“Parties should take particular measures concerning the depiction of tobacco in entertainment media products, including requiring certification that no benefits have been received for any tobacco depictions, prohibiting the use of identifiable tobacco brands or imagery, requiring anti-tobacco advertisements and implementing a ratings or classification system that takes tobacco depictions into account.”

International streamers already comply with all these requirements, which explains why they have been able to continue to do business in all the countries that have signed on to the Convention, including the US which, although it has not become a Party, applies similar conditions.

The Indonesian regulation that would require streaming services to remove all depictions of smoking is excessive. While the goal of reducing consumption of tobacco products is laudable, to require the complete elimination of depiction, especially from streaming services, is using a sledgehammer to kill a fly, destroying the furniture on which the fly landed in the process. Appropriate, targetted regulation deals directly with the problem, avoiding unintended consequences and collateral damage. Regrettably, the current Indonesia draft regulations do not do that. By all means, fight tobacco consumption and stop the tobacco companies from finding loopholes. But do it in a way that is effective and reasonable.

© Hugh Stephens, 2024. All Rights Reserved.

Distasteful Yes, But Not Copyright Infringement: Publishing “Basic Fact” Unauthorized Obituaries is Going Strong (And Often Getting it Wrong)

Image: Shutterstock.com (with AI Assist)

Tempus fugit. Time flies, as we all know. It’s hard for me to believe that it was over 5 years ago that I wrote a couple of blogs on obituary piracy;  (The Deadliest Aspects of Copyright and Obituary Piracy Punished: Has Infringement No Bounds?) Those blogs were about an outfit in Canada that styled itself Afterlife. Afterlife’s business model was to post obituaries on their site they had harvested from newspapers and funeral home websites in order to sell virtual candles, flowers and other forms of remembrance for the departed. Not surprisingly, families of those so featured objected since they had not authorized Afterlife to publish, let alone monetize, their personal tributes to their deceased family members. A class action lawsuit was launched, alleging copyright infringement since an obituary, a creative work embodying original expression, can be protected by copyright. It is the story of a person’s life, often written by a close relative or in some cases in advance by the deceased person themself. In the Afterlife case, the Federal Court of Canada awarded damages of $20 million and issued an injunction preventing Afterlife from continuing to publish.

In this case copyright infringement was the legal tool used to stop Afterlife’s activities. While the court case succeeded in getting Afterlife to stop what the judge called “obituary piracy”, those running the business simply changed tactics. They declared bankruptcy (no doubt to thwart any collection of damages) and opened a new site called Everhere. Everhere avoided copyright infringement by highlighting a death, but then linking to the obituary posted elsewhere without copying or reproducing it. Among other things, it became a specialized ad-supported obituary search engine. I lost sight of Everhere and its somewhat sleazy business model until unauthorized obituaries again hit the headlines in a recent story published by the Canadian Press.

This time the company is known as Echovita. At least some of the same people involved with Afterlife are apparently running Echovita, according to the Canadian Press report. Apart from the fact that families are upset that the passing of a loved one is being monetized by an unknown third party on its website, often to the surprise of friends and relatives who may have searched for the obituary on a search engine and not realized that the Echovita obituary is an unauthorized version, it appears that not infrequently mistakes are made. This is because in order to avoid its previous problems with copyright infringement, Echovita is doing quick rewrites of material it has skimmed from the internet. No doubt it is paying various freelancers by the piece to produce obituaries from so-called “tombstone data” (the basic facts of a person’s life), and the mass production of large numbers of these each day naturally leads to mistakes being made. The author of the Echovita obit doesn’t know the deceased from Adam. They don’t know if he or she had two or ten grandchildren, or none and, to be blunt, they don’t care. The obituary is just another commodity they are producing. They could just as easily be writing about soap flakes or breakfast cereal. In defending itself against accusations that it made mistakes in some obituaries, Echovita is reported as saying it regrets that it made a “human error” when taking the “basic facts” from an obituary posted on the internet, while noting that it was legal to do so.

Going to the Echovita site is instructive. Obituaries are not restricted to Canada. People using the website have the option of purchasing virtual candles that will “burn” for various periods from a week to a year, depending on payment and preference. The virtual candle will “burn” on the site of the barebones obit published by Echovita. You can also plant a tree or order flowers in remembrance. And there are ads for various things, such as Ancestry.com. It makes my skin crawl.

From a legal perspective (at least with regard to copyright infringement), Echovita’s claim that what they are doing is legal appears to be correct.  Facts and ideas cannot be protected, only the expression of an idea based on those facts or ideas. The same is as true for obituaries as it is with news. While the New York Times can copyright its reporting of, say, the war in Gaza, anyone can report on what is happening in Gaza. No news outlet has a monopoly on the news although they can protect their expression of the news, an issue I examined in this blog a couple of years ago, (Do News Publishers “Own” the News? (And Should They be Compensated when Others use News Content they Publish?). Distasteful as it seems, there is no law that I am aware of that stops Echovita or anyone else from taking basic data (facts) from a published obituary (name, date of death, place of service, etc) and rewriting and publishing it. Using it for commercial purposes might be something else. I wonder if the terms of service of a funeral home website might limit the lifting and reusing of this information for commercial purposes, but I will leave that idea for the lawyers to ponder.

According to the Canadian Press article, the president of the Funeral Services Association of Canada is lobbying the federal government to strengthen privacy legislation to deal with the republication of obituary details, but has had little response. I am no expert on privacy legislation, but I know it is a minefield where the need for personal privacy must be balanced against freedom of expression. In the meantime, the best solution the Funeral Association can offer is to contact sites like Echovita and ask that an unauthorized obituary be taken down, although they would have no incentive or legal obligation to comply.

Dealing with obituary piracy, whether officially copyright infringing or not, is a difficult issue. Undertaking reform will not be simple. Pardon the pun.

© Hugh Stephens, 2024. All Rights Reserved.