Month: March 2025

Copyright, Cultural Issues and Canada’s General Election, 2025

Image: Shutterstock (AI generated)

As we complete the first few days in what is the shortest election campaign in Canadian history, the minimum 37 days required by law, where do the copyright and cultural industries stand with respect to electoral platforms and public consciousness? Given the overwhelming focus on dealing with economic and even potential political disruption coming from south of the border, along with traditional bread and butter issues like the cost of living, especially food and housing, one could be tempted to say that cultural and copyright issues are largely invisible. Party platforms have not yet been released (and are probably still being worked on) and by the time they are made public, the election will be well underway. So while there still may be a couple of small references to copyright issues in party platforms (as occurred in the 2021 election, none of which led to any substantive legislation), they will simply be part of a laundry list of possible actions in many disparate areas. However, that has not stopped the cultural sector from outlining its policy proposals, which have been laid out articulately by the Coalition for the Diversity of Cultural Expressions (CDCE), an umbrella group that represents more than 350,000 creators and artists, and more than 3,000 cultural enterprises. Despite the fact that copyright issues are not at or even near the top of the agenda, there is a strong undercurrent of Canadian nationalism in this election that will inevitably have an influence on policies in the cultural sector.

In 2021 the governing Trudeau Liberals included a promise to “protect Canadian artists, creators and copyright holders by making changes to the Copyright Act including amending the Act to allow resale rights for artists”. They were re-elected but did nothing. The Conservatives for their part undertook “recognize and correct the adverse economic impact for creators and publishers from the uncompensated use of their works…”. They weren’t elected so the commitment was meaningless. This time proposed changes to copyright legislation are unlikely to move the needle for any party although the issue of the unauthorized use of copyrighted content to train AI still needs to be resolved, since AI will become a front-burner issue for any party elected. The CDCE’s paper addresses this issue, among others, in its 9 recommendations. Broken down into 4 buckets, the CDCE’s proposals address (1) International Trade and Cultural Sovereignty (2) Broadcasting and CBC/Radio Canada (3) Copyright and (4) Artificial Intelligence and Culture.

The CDCE proposal under “International Trade” is to insist that the cultural exemption clause be retained if the CUSMA/USMCA is renegotiated, and that cultural activities, goods and services be excluded from all future agreements. The cultural exemption clause, (Article 32.6 of the CUSMA) is based on a similar exemption in NAFTA and the original US-Canada bilateral trade agreement of 1989 but is more of a political fig-leaf than a real protection since if the provision is invoked, the US can retaliate with equivalent effect in any trade sector. However, it provided comfort to the cultural sector at a time when free trade with the US was seen to make Canada vulnerable culturally. Thirty plus years of bilateral, and now trilateral, trade proved that fear to be unfounded—until now—and the cultural exemption has never been used. During the period from 1989 to the present, even through the ups and downs of Trump 1.0, the fundamentals of the initial bilateral Free Trade Agreement, then NAFTA, and now the CUSMA/USMCA were basically respected by all parties. Under Trump 2.0 this has all been called into question. If the Trump Administration is going to disavow the basic elements of the CUSMA, having a cultural exemption clause becomes less than meaningless.

On April 2, the US will unveil its “reciprocal tariff” regime. It has arrogated to itself the right to include, in addition to tariffs imposed by other countries, self identified non-tariff measures in its calculations. Among these may be various cultural support measures imposed by Canada on foreign entities operating in Canada requiring them to make financial contributions to Canadian content. If that happens, the US will be violating yet again the provisions of the CUSMA/USMCA as it has already done with regard to the imposition of tariffs on some products on the specious grounds of fentanyl trafficking from Canada to the US, (less than 20kg in all of 2024). However, given the surge in Canadian nationalism as a result of the tariff threats but more particularly the verbal diarrhea coming daily from President Trump about Canada becoming the 51st state, it is unlikely that any Canadian government would throw Canada’s cultural identity under the bus for the sake of preserving tariff-free access to the US market for some commodities. Thus, seeing Canada sacrifice cultural support measures that may annoy some US businesses operating in Canada (like online streaming content providers) in return for a degree of tariff relief is an unlikely outcome in the present circumstances.

This surge of nationalism relates to the second of the CDCE’s “demands”, protecting the CBC and the Canadian broadcasting environment. Ever since Pierre Poilievre became leader of the opposition Conservative Party, one of the Party’s mantras has been “defund the CBC”. There is no question that the CBC business model is in need of reform, particularly its English language entertainment television service which captures a very small market share, but CBC radio, CBC news broadcasts and CBC’s French language service, Radio-Canada, remain highly relevant, as this CBC explainer attempts to show. Given the need to protect national identity in the face of the Trumpian onslaught, and the recent rediscovery that perhaps Canada is not so “broken” after all, if ever there was a need for this national institution, it is now.

The third basket of issues raised in the CDCE position paper relates to copyright concerns, which get very little traction among the general electorate but are important to the creative and cultural community. Once again, the CDCE reminds parties of the lack of an Artists Resale Right in Canada (noting previous promises to establish this measure), as well as some other longstanding issues like fair remuneration for writers and publishers for the use of their works in the education sector and extending the private copying regime to electronic devices. This would impose a small levy (about $3) paid by manufacturers and embedded in the cost of a smartphone to compensate for unregulated widespread copying of music on these devices, with the funds flowing back to music creators.

The final bucket deals with Artificial Intelligence (AI) and copyrighted content. At the present time there are some 40 lawsuits in the US pitting rightsholders against AI developers, and even a couple of cases in Canada. Canada has been slow off the mark in addressing this issue; at the moment there is no Text and Data Mining exception in Canadian copyright law and both rightsholders and AI developers are not clear on the ground rules. The CDCE is asking that a legislative framework be adopted that includes the key principles of (1) Authorization (by the rightsholder) (2) Remuneration (payment for use of copyrighted content) and (3) Transparency (the establishment of disclosure rules as to what training data is used in AI systems and ensuring that all AI-generated content is clearly identified). These are reasonable asks but there is no guarantee they will be respected.

In the US, AI developers are pushing the Trump Administration to give them a pass on respecting author’s copyright, notwithstanding the cases before the courts, using the argument that the US will lose the AI race to China if US developers cannot help themselves freely to the content of others. OpenAI (which is being sued by the New York Times) and Google argued in submissions to the US government that giving them unfettered access to data, including content owned by others, is essential for national security. Described by blogger David Newhoff as “tech bro bombast”, OpenAI’s attempt to wrap itself in the national security blanket is a cynical ploy to get around the inconvenient fact that it and other AI developers are hijacking the creative work of authors, artists, and musicians without permission or compensation while creating outputs that in a number of cases can compete with or even displace the original works that contributed to their training. A similar situation is developing in the UK where the creative community is pushing back against the original copyright carte blanche that the UK government seemed inclined to give to the tech community, in the name of AI competitiveness. Canadian governments are not beyond succumbing to the siren calls of the AI community and it is timely to establish some guiding principles, of which Authorization, Remuneration and Transparency are a good place to start.

However, while AI and copyright are not going to become election issues, national identity, which is closely intertwined with cultural sovereignty, surely is. Indirectly, copyright will be important as it is one of the foundation stones of cultural sovereignty, an issue that would have played second fiddle to economic issues like food inflation, carbon pricing, cost of housing, fuel and utility costs etc until Donald Trump started spouting his annexationist nonsense.

Frankly, had Trump really wanted to absorb Canada (eventually) he should have brought Canada inside the US economic tent and made the country even more reliant on the US market, by providing it with an exception to his attempts to take on the world trading system. Instead, he has woken Canadians from a restful, dependent slumber brought on by three decades of relatively uncontroversial free trade and economic integration and made them realize that they have no one to depend on but themselves. In doing so, he has revitalized a sense of nationalism that will play out in this election. Who can best defend Canadian interests has become the litmus test for Canadian voters, leading to a remarkable resurgence for the Liberal Party under new leader Mark Carney after the political corpse of Justin Trudeau was removed from the electoral scene. This may or may not change during the course of this short campaign. One thing is certain; while copyright issues per se will not get much profile, cultural identity issues will certainly be in the spotlight. This is a shift in emphasis that in the long run is likely to benefit the creative sector.

© Hugh Stephens, 2025. All Rights Reserved.

Donald Trump’s Tariff Threats: Their Potential Impact on Canada’s Cultural Industries

Image: Shutterstock.com

With a general election in Canada now set for April 28, attention will be focussed south of the border to see what Donald Trump says and does next. Apart from his tiresome and insulting trope about Canada becoming the 51st US state, how best to deal with the economic fallout from the imposition of unilateral US tariffs on Canadian exports to the US will be the big election issue. Indeed, the drumbeat of tariff threats emanating from self-proclaimed “Tariff Man” is becoming overwhelming, both in terms of tariffs already applied, but also regarding potential future tariffs. As we have already seen, the uncertainty and almost daily changes, (government by tweet), are roiling markets and undermining investor confidence. With respect to Canada there have been repeated threats of what is to come while some tariffs, such as those on steel and aluminum that were applied globally, are already in force. Then there are the threatened 25% tariffs on all Canadian (and Mexican) imports, except for energy products which will be taxed at a 10% level, imposing additional costs on US consumers. (The example of potash, an essential product needed by American farmers is an interesting case study. It is basically only available from Canada, unless you import it from Russia, Belarus or China. The US does produce a small amount but 85% of US potash consumption comes from Canada. So much for President Trump’s mantra that Canada has nothing the US needs. When US farmers squealed loudly, the duties on potash were suddenly lowered from 25% to 10% and then suspended completely under an exemption for all products covered by CUSMA).

The 25% tariffs designed to hinder the export of automobiles and car parts (amongst other products) manufactured in Canada from being shipped to the US —a measure which incidentally contravenes the terms of the US-Canada-Mexico Agreement (USMCA/CUSMA)—are temporarily on suspension given the representations made by US auto manufacturers who had to explain to the White House how integrated North American supply chains work, but any products not covered by USMCA/CUSMA are still subject to the 25% tariff. The pretext for this violation of a ratified trilateral trade agreement is supposedly the “national emergency” created by the flow of fentanyl and illegal immigrants from Mexico and Canada. The only problem with this rationale is that, in the case of Canada,  there is a greater flow of illegals from the US to Canada than vice versa, and the seizures of fentanyl at the northern border by US officials in 2024 totalled less than 20 kilos, less than one percent of the amount seized on the southern border. Thirteen grams (that’s less than half an ounce) were seized in January. This year, US border officials have caught more people smuggling eggs from Canada into the US (where the price of eggs has shot up owing to avian influenza in US poultry flocks) than fentanyl. But the facts appear irrelevant to the Trump Administration; what is important is to create a pretext to violate the USMCA.

That pretext was used to trigger the International Emergency Economic Powers Act (IEEPA). This legislation allows the President “to deal with any unusual and extraordinary threat, which has its source in whole or substantial part outside the United States, to the national security, foreign policy, or economy of the United States”. It was first enacted in 1977 and is designed to deal with acts of terrorism or other threats to the security of the United States. It confers wide, albeit temporary, powers on the Executive Branch and has been used in situations like the Iran hostage crisis in 1979, the Soviet invasion of Afghanistan and to deal with various identified terrorist groups. Using it to punish Canada because under 20 kg (43 lbs). of fentanyl were seized in the course of a year at the Canada-US border is clearly an abuse of the intent of the Act. Notwithstanding, that is what Trump used to impose USMCA noncompliant tariffs on Canada.

The temporary suspension of the 25% tariffs on Canadian (and Mexican) imports will apparently end on April 2, when Trump plans to impose reciprocal tariffs on a global basis. How these will be calculated is anyone’s guess. The President has indicated that in addition to whatever tariff irritant he can find, he might also include other measures in US calculations that in his view discriminate against US goods and services. Thus, while most US products enter Canada tariff free (with the notable and unfortunate exception of most dairy products, which are subject to Canada’s outdated supply management system), Trump could take aim at other policies he doesn’t like. For example, while many US banks operate in Canada, none of them are full-service retail banks allowed to take deposits (but nor are they required to have the same capital requirements). Then there is the fact that Canada, like the EU, imposes a value-added tax (VAT) on most products (basic foodstuffs being the primary exception), called the GST (Goods and Services Tax). This is another potential target even though it is applied without discrimination to US, Canadian or products from any other country. Likewise, longstanding measures that provide protection and subsidies for Canadian cultural industries, like broadcasting (AV and music) content quotas or more recent mandatory financial contributions to Canadian content funds, along with funding obligations to support local journalism, could potentially become targets.

Would Google try to reopen the commitments it finally made to support Canadian journalism in order to avoid designation under the Online News Act? Will the mandatory “contributions” to Canadian content creation that the CRTC has imposed on foreign streamers become an issue? A prominent US trade association, the Computer and Communications Industry Association (CCIA), went so far as to claim that ”the CRTC’s structure of mandatory contributions contravenes Canada’s commitments to the United States under CUSMA”. While I think that claim is doubtful, if the Trump Administration regards the CUSMA as just a piece of paper to be ignored at will, US industries should think twice about using it as a lever against Canada. In any event, in my view the best approach is to continue to stress the value of cooperation and mutual benefit, as Canada has been trying to do by explaining to the Trump Administration why tariffs are self-defeating. In terms of AV production, the contribution that US content producers make to the Canadian production industry is significant even if there are disagreements about the extent to which US production in Canada helps or hinders creation and distribution of Canadian content.

It is important to note that all countries impose investment or trade restrictions of one sort or another, and the US is no exception. These restrictions are weighed in terms of the balance of reciprocal benefits when trade agreements are negotiated, including the current USMCA/CUSMA signed by Trump himself in his first term. But if you are not inclined to respect the commitments you have made, and intend to ignore carefully negotiated and signed treaties, then any domestic measure can become a target. Uncertainty as to what could happen next is a major concern. Two Canadian cultural industries that are keeping their heads down and hoping for the best are art dealers and book publishing.

Earlier this month, the Globe and Mail reported that art dealers and galleries are facing slowdowns in the face of the uncertainty brought about by the Trump tariff threats. Books, art and other informational materials were granted an exemption when Trump first imposed the tariffs on Canada, using the excuse of fentanyl trafficking. Buried within the legislation used to suspend USMCA/CUSMA obligations, (the IEEPA referred to in paragraph 3 above) is a provision that creates certain exceptions, amongst which is “any information or informational materials, including but not limited to, publications, films, posters, phonograph records, photographs, microfilms, microfiche, tapes, compact disks, CD ROMs, artworks, and news wire feeds”, unless controlled by some other authority. (Section 1702 (b)(3)). While US Customs noted the exception when publishing its Notice of Implementation, the on-again/off-again tariff implementation has created anxiety and uncertainty, not least of which is the possibility that any random Customs officer can hold up a shipment based on an individual (mis)interpretation of the regulations. Compounding the issue is the announcement by Canada of 25% retaliatory tariffs that include, among other things (the targets of the retaliation are wide covering everything from toilet paper to drones), “Paintings, drawings and pastels, executed entirely by hand”.

Book publishers are also exempted under the IEEPA and are keeping their heads down, as noted by another article in the Globe. Many Canadian publishers do not ship much to the US but some do, including companies that are exclusively printers rather than full service publishers. In the case of Friesens Corp, a printer in Manitoba, the bulk of their business is from US customers. However, now a new threat has risen for Canada’s independent book sellers. Books have been included on Canada’s retaliation list, and if books from the US are subjected to a 25% retaliatory tariff, the cost will be passed on to bookstores, and ultimately consumers. Independent bookstores already work on very thin margins and an additional charge will likely affect sales. Harm to Canadian business and consumers is the flipside of punishing US exporters, just as harm to US consumers will result from US tariffs on imports. Surely it would be best to leave a cultural product like books out of the trade war.

What happens next with regard to tariffs on exports to the US, from Canada or elsewhere, seems to depend on Donald Trump’s mood of the day. The expected announcement of “reciprocal tariffs” on April 2 will create further uncertainty and likely retaliation, further feeding the spiralling trade war. The fact that import tariffs are levied on the importer and are largely passed on to consumers seems not to have registered with the Trump Administration. They can certainly raise revenues, but if the end goal is to impede imports so that all production is reshored to the US, then presumably the revenue windfall (largely ultimately paid for by US consumers) will ultimately disappear. To depend on tariff revenues to fund more tax cuts in the US is ultimately a self-defeating strategy. In the meantime, the US economy will have suffered the impact of increased prices on all imported goods.

The Trump tariffs have had the effect of causing maximum disruption and chaos, and if that was the goal, then Donald Trump has succeeded. In the meantime, Canadians have until April 28 to figure out which political party and leader is best equipped to help navigate the treacherous waters ahead. Whatever happens, copyright and cultural industries are unlikely to escape getting wet.

© Hugh Stephens, 2025. All Rights Reserved.

This post has been updated to include reference to potential Canadian retaliatory tariffs on US book imports and the impact this will have on independent bookshops in Canada.

How AI Can Destroy Local Journalism

Image: Shutterstock.com

We all know that local journalism is under extreme pressure. Long established regional newspapers are closing or are being turned into little more than franchise operations where a bare bones local newsroom contributes a modicum of local news to a newspaper fleshed out with filler from national wire services or mother publications. The regional titles in Canada owned by PostMedia are a prime example of this phenomenon. Some digital startups have helped fill the gap, but they too are struggling. There seems to be a reluctance on the part of many to pay for news through subscriptions, while small online publications are forced to compete with everyone from Google and Facebook on down for ad dollars.

One of the supposed remedies, at least in Canada, has been to create various funds to support local journalism. There are a range of programs including the government funded Local Journalism Initiative, launched in 2019 to encourage local news production in “news deserts” and underserved communities, administered by News Media Canada (an industry group), tax credits to offset journalist salaries if the organization is a “Qualified Canadian Journalism Organization” (QCJO), and most recently the Google-funded $100 million fund for media outlets. This was established as a result of Google’s deal with the government to get a five year exemption from being designated under the Online News Act. According to recent reports, the Google fund’s annual contribution to a journalist’s salary will be in the range of C$13,000 to C$20,000 depending on the number of applying journalistic enterprises deemed to qualify.

All these Band-aid measures are designed to staunch the loss of print and digital publications, which has created news deserts in many parts of Canada and the US. In a 2023 report from Northwestern University’s School of Journalism, as reported by Forbes, it was estimated that almost 3000 print newspapers out of approximately 9000 in the US had ceased publication since 2005. The average loss of newspapers in 2023 was 2.5 per week. In my home province of British Columbia alone, Global News reported that the number of daily papers dropped from 36 to just 13 in the six years from 2010 to 2016. According to the Local News Research Project housed at Toronto Metropolitan University’s School of Journalism, over 500 local radio, TV, print and online news operations shuttered in 345 communities across Canada since 2008 although during that time, around 200 new local news outlets, many of them exclusively digital offerings, launched in 152 communities. However, just one opened in 2023. The decline of local news in Canada was well documented in a study released in February of this year by the Public Policy Forum, “The Lost Estate”. The study examines a number of possible remedies, including philanthropic engagement, community foundations, better targetting of government support programs and increased government advertising in local media.

The reasons for the decline of local news are many; the rise of social media, the migration of ad revenues to giant online platforms like Google and META, the reluctance of a younger generation of consumers to pay for news (especially digitally provided content), unauthorized password sharing (even by government!), rising costs of print, and so on. Newly launched digital outlets have tried to fill the gap, but they are facing challenges in getting sufficient ad revenue or subscriptions. In Canada, some will qualify for tax credits or funding from Google, but it will depend on whether they employ at least 2 full time journalists. For many of these digital outlets, their principal modus operandi is to aggregate content from other sources, provide a quick rewrite summary and then insert a link to the original source, thus avoiding copyright infringement. There is some but not a lot of original local journalism.

I have no doubt that with the growing use of AI, some of this initial screening is done through use of artificial intelligence. AI may even be being used to create summaries and rewrites. This saves time and money—but unfortunately cuts down on the need for real journalists. The evaluation of AI’s utility in local journalism is typical of its use in many other areas, from screen writing to auditing to medical diagnoses. It can be a useful tool to enhance productivity, but often at a cost somewhere else, such as with respect to employment. Even if an experienced employee can employ AI to enhance what they are working on, AI could eliminate the beginner or training jobs that help develop the required experience to do this. These challenges are not new and not unique to local journalism.

What is new is the use of AI to take the aggregation model followed by many small local online journals a step further and go nation-wide, a development discovered and highlighted by the Neiman Lab. The Neiman Lab is part of the Neiman Foundation for Journalism, established in 1938 at Harvard. It administers what is proclaimed to be the oldest fellowship program in the world for journalists. The Foundation also publishes a quarterly magazine, Neiman Reports, dedicated to a critical examination of journalism and other journalism-related programs. In an excellent piece of investigative sleuthing, the Lab discovered that “Good Daily”, which operates in 47 states and 355 towns and cities across the US, targeting small town America, is run by just one person, Mathew Henderson, armed with an AI program. If this seems hard to believe, read on.

According to Neiman’s investigation, Henderson operates his “media empire” out of New York. He uses an AI bot to scan the news daily in each local market. The AI program curates the most relevant stories, summarizes them, edits and approves the copy, formats it into a newsletter, and publishes it. The same day! Readers in these 355 towns are led to believe that this is a local publication. It has local testimonials, although the same testimonials, slightly tweaked, appear in various editions around the country. The publications also share the same “About” information and the same mission, which is “to make local news more accessible and highlight extraordinary people in our community.” Henderson claims his automated newsletter is actually helping local publications by driving traffic to them. This is the same argument put forward by META when refusing to pay for news content that it uses on its platform to attract and retain viewers (and sell ads).

Henderson’s business model is to sell advertising and solicit readers for donations. The advertising pie is not infinite, so it is obvious that his “local” newsletters are just one more source of competition for local media chasing ad dollars. Just to be clear, there is nothing illegal about what Henderson is doing. Aggregating content and linking to it is not a violation of copyright law. The lack of full disclosure is a bit disconcerting but I am doubtful if there is anything illegal about having a corporate veil. What is problematic is the cannibalistic nature of his business model, enabled by AI.

This kind of operation, fuelled by AI, can only operate if there is local content to aggregate. The Good Day publications contribute nothing, absolutely zero, to content creation. They, like Facebook, are the ultimate cannibalistic free riders. They can continue to operate successfully, free riding on content created by others, so long as those “others” remain in the business of producing content. However, the more successful businesses like Good Day become, the less viable will be the local journalism sector that produces the content its free-riding competitor subsists on. In the end, the result will be an ouroboros. (Great Scrabble word by the way). The AI driven aggregator will in the end devour the very source of its content. That may be a few years down the road, but logically that is what will happen. It’s like eating your seed grain.

Is there a solution? Many remedies have been proposed, but if there is a silver bullet, I don’t know what it is. For many in the media, finding a way to get the big platforms that benefit from media content to contribute financially to journalism is one avenue, but as we have seen in Canada, Australia, and California, the platforms will pull out all the stops to prevent being required to do so, especially META, which has thumbed its nose at Canada and, now, Australia.

Government subsidies, such as Canada’s Local Journalism Initiative, are resisted by many journalists lest the industry become beholden to government handouts. Holding government to account is one of the key functions of the media, the so-called “Fifth Estate”. Can a subsidized media be trusted to do so? On the other hand, without subsidies will there be any viable local media left? Maintaining its independence is one reason why the small Ottawa-based online publication, Blacklock’s Reporter, that is locked in a David and Goliath struggle with the Government of Canada over the government’s abuse of password-sharing, is opposed to initiatives like the Online News Act. (For the record, that is their view, not mine, but it is a position I respect). Potentially another option could be a tweaking of tax laws to encourage businesses to place ads on local media rather than with the giant international platforms, but in the end business will and should be able to spend its ad dollars where it believes it will get the best results.

For every potential remedy to the problem of keeping local journalism alive, there is a potential downside, just as there is with AI generally. For all its potential advantages AI also has the potential to destroy journalism. Good Daily may be the canary in the coalmine, a fully legal but particularly egregious use of AI putting yet another nail in the coffin of local journalism.

© Hugh Stephens, 2025. All Rights Reserved.

Using Copyrighted Content to Train AI: Can Licensing Bridge the Gap?

Image: Shutterstock

The struggle between authors (writers, artists, musicians) and AI developers over the unauthorized and uncompensated use of copyrighted works to train AI applications continues, both in the courts (here is a summary of the current state of play in the US where most of the litigation is taking place) and in the political arena, such as the UK government’s latest initiative to put its thumb on the scale in favour of the AI industry, now slowed down by opposition within Parliament. The creative industries in Britain are still nervous, however, as demonstrated by the coordinated “Make it Fair” campaign organized by leading UK newspapers on February 25. While the courts may provide some guidance, it is unlikely to be dispositive and is almost certainly to be somewhat contradictory and lengthy, given the appeal process that will play out. With new applications being rolled out every day, AI appears to be unstoppable. Let’s accept that this is the case. If so, what then will be the rules governing the use of AI training content, particularly content that is protected by copyright, such as books, journalistic output, paintings, musical compositions etc.?

It is already apparent that at least some of the AI output trained on these materials will compete in the marketplace with the original works. If that is the case, then surely some of that additional value should be shared with those who helped create the content initially. The way this will most likely be done is through licensing in the form of payment and permission for use of the copyrighted creative output that enabled the training to take place. Licensing would also help resolve another potential issue, the possibility that the final product produced by the AI algorithm infringes on the copyright of the works on which it was trained. This is unlikely to happen in the case of written works but is certainly potentially possible with graphic or musical works.

While many have called for licensing as a solution, there are many challenges to be overcome to make it work effectively. Yet some licensing is already taking place between AI developers and owners of well delineated data sets. As an example, various newspaper and magazine publishers have already reached licensing agreements with AI providers. OpenAI has signed licensing deals with the Wall Street Journal, Times of London, the Financial Times, Time, Le Monde, Axel Springer and others. This is in marked contrast to OpenAI’s relationship with the New York Times, which has led to one of the most prominent lawsuits in the field, with the Times suing OpenAI for copyright infringement. The reason for this lawsuit, of course, is because licensing negotiations between the two entities broke down. Some photo and image licensing companies have concluded AI deals (Shutterstock is the most prominent example) while others, such as Getty Images have not. (Getty is suing StabilityAI in the UK). Eventually most of the institutional or corporate holders of valuable content in one form or another will likely reach, or attempt to reach, licensing deals with the major AI developers. But that still leaves out an awful lot of copyright-protected content.

The conundrum is how to deal with the millions of individual creators who produce content in different formats, and tie them into a workable licensing regime. The first challenge is how to even figure out who is producing content that is likely to be used by AI developers. The second is to calculate how much that use is worth. Then there is the challenge of how to administer a collective licensing scheme in a way that is both practical and affordable and where the small amount of royalties for individual works are not swamped by the administrative costs of collection and disbursement. Finally, there is the question of how to resolve the issue of competing licensing organizations in order to provide more or less one-stop-shopping for the AI industry.

It is worth noting that one-stop-shopping currently does not exist in any area of collective licensing. Different collectives represent creators in different fields so music, publishing, art, broadcasting and visual arts licensing are all represented by different organizations, in some cases with more than one collective in a given field. The Copyright Board of Canada lists 36 copyright collectives on its website. I haven’t seen a definitive list for the US but this university website lists about the same number.

Whereas users of music only have to deal with a handful of CMOs (collective management organizations), and users of text based content (online or offline) need only to acquire a reprographic license from the major licensing collectives for published works, such as the Copyright Clearance Center in the US or Access Copyright in Canada, AI developers access the full gamut of content. It will be challenging to make access easy for the AI development industry, a point developed by Dr. Pamela Samuelson of the University of California, Berkeley, well-known copyright scholar (and skeptic, let it be added). In her recent paper in the UCLA Law Review (“Fair Use Defenses in Disruptive Technology Cases”), Samuelson focuses primarily on the question of fair use—as suggested by the title—but also examines the issue of a collective licensing regime for generative AI development. She manages to raise just about every objection conceivable (see pp.80-86 of the document for more details);

–generative AI uses all forms of content therefore the licence would have to be very broad
-an issue would arise as to whether content for training was used just once, or on repeat occasions
-it would be very difficult and costly to administer given that there could be literally billions of creators involved
-creators would get very little revenue; the bulk would go to the administering agencies, the CMOs
-it would be difficult to determine value and to set a price on each transaction
-what about orphan works?
-differing national regimes might create confusion; alternatively some countries might not require a licence payment, giving them an unfair advantage
-it would be unfair to startups since the incumbents have already scooped volumes of content without payment.

She notes that creators may lose out, but since AI will affect the livelihoods of so many others, this is not exclusively a copyright problem. Tough luck creators.

Clearly Dr. Samuelson is not in favour of a collective licensing regime for content appropriated by AI developers, yet despite her firehose of cold water, there are a number of promising developments in this area. For example, the Copyright Clearance Center (CCC) in the US recently announced it would provide AI re-use rights within its Annual Copyright Licenses, making the CCC’s licence “the first-ever collective licensing solution for the internal use of copyrighted materials in AI systems.” Note the caveat. While covering re-use of content for AI applications, the CCC makes it clear that;

“The license enables participating rightsholders to fulfill the needs of companies that require an efficient way to legally acquire the rights to use copyrighted materials within AI systems for internal use.”

Not training. The Copyright Agency in Australia has done something very similar.

“Starting from February 2025, Copyright Agency will extend its Annual Business Licence to cover staff of licensed businesses who include third party material in prompts for AI tools (and) copy and share outputs from AI tools with colleagues”.

However, it does not apply to AI training and does not allow capture of the content outside the business, such as by an externally provided AI tool.

Likewise, the Copyright Licensing Agency (CLA) in the UK issues a Text and Data Mining (TDM) Licence. The CLA’s website explains that TDM “is the process of transforming unstructured content into a structured format to analyse, extract and identify meaningful information and insights. By using TDM, organisations can harness the power of vast volumes of information and data, capturing and revealing key concepts, trends, and hidden relationships.” Sounds quite a bit like training generative AI, but it’s not.

“CLA’s TDM licence extension includes rights covering use of published content for TDM purposes. This does not cover the use of content in training or prompting Generative AI models.”

Canada’s equivalent CMO, Access Copyright, is actively examining the issue, as it notes in its new strategic plan for 2025-2028;

“Like collective rights management organizations around the world, we will actively explore how we might enhance our corporate licence offerings to include uses related to AI, providing Canadian rights holders who wish to participate in the emerging market for AI licensing to do so, either in Canada or by virtue of reciprocal agreements with sister organizations.”

It is clear that these Reproduction Rights Organizations (CMOs by another name) are cautiously feeling their way forward to find the appropriate role for collective licensing. Meanwhile the private sector has not been sitting idly by. Forbes reports that so many content aggregation startups have been established that they have formed a Data Providers Alliance. Recently launched “Created by Humans” is another commercial entrant that is pitching itself to authors.

“Take control of your work’s AI Rights and get compensated for its use by AI companies.”

As these new enterprises enter the market, it threatens to become quite crowded. Just as there are more and more AI companies, including new entrants like DeepSeek, a proliferation of new sector-specific content-aggregators will make licensing more challenging. If the CMOs wait too long, they will face entrenched competition. Not all these new aggregators will survive. In the end, AI developers will not subscribe to multiple content licensors; they will go with the ones that provide the broadest coverage. It will be a Darwinian selection process.

While this is happening, other countries are experimenting with the concept of extended collective licensing for AI content. This allows CMOs to grant licenses on behalf of both their members and non-members alike. An extended collective licence is not a compulsory licence but it could lead to such a system being established. Spain was first out of the gate, but has since pulled back after the proposed Royal Decree attracted the criticism from many rights holders that it would proscribe their options. Yet it is one of many solutions being tested.

The recently released US Copyright Office report “Identifying the Economic Implications of Artificial Intelligence for Copyright Policy” includes an extensive discussion of licensing possibilities, including examining the pros and cons of a new statutory blanket licence. This would need to include a provision excluding rightsholders (such as entities that have already reached licensing agreements with AI developers) who have the ability and wish to issue voluntary licences that generate greater remuneration than a statutory payout would earn. This raises thorny opt-in/opt-out issues. Compromises will be required, but the challenges are not insurmountable.

The trick is to devise a system that will capture as much content as possible while allowing some flexibility to rightsholders, allocating payments in way that is fair and efficient (the USCO paper suggests that revenues associated with a work could serve as a rough proxy for its relative value), at the same time minimizing administrative costs so that expenses do not exceed potential revenues for rightsholders holding limited content inventory. Can it be done?

Despite the many obstacles identified by Dr. Samuelson and others, I am convinced that in the end collective licensing for content used in AI development and applications will become as accepted as the collective licensing regimes for use of various forms of copyrighted content today. The way forward won’t be straightforward; there will be zigs and zags. The courts and legislatures will play a role, as will authors, publishers, and the AI developers themselves. But in the end we will get there. Licensing, including some form of collective licensing, is the inevitable bridge that will bring AI developers and copyright holders together.

© Hugh Stephens, 2025. All Rights Reserved.