Anthropic's record copyright deal stalls over fee questions

What happened
A proposed record-breaking US$1.5 billion copyright settlement involving AI giant Anthropic and a substantial class of authors has hit a snag. A US federal judge, Araceli Martinez-Olguin, declined to grant final approval for the deal, seeking further clarification on attorney fees and payments designated for lead plaintiffs. This judicial intervention means over 120,000 authors and copyright holders in the United States must continue to wait for resolution in a case that dates back to 2024.
Anthropic faced allegations of illicitly downloading over seven million pirated books from so-called 'shadow libraries' — LibGen and PiLiMi — to train its artificial intelligence models. While a now-retired Judge, William Alsup, previously ruled in June 2025 that using these books for AI training could be considered 'fair use' if obtained legally, he simultaneously found that collecting pirated copies in a 'central library' for purposes beyond mere training was not protected under fair use provisions. This nuanced ruling initially set the stage for a trial where Anthropic could have faced hundreds of billions of dollars in damages.
To avert a potentially catastrophic trial, Anthropic agreed to the US$1.5 billion settlement, which received preliminary approval in September 2025. The case was subsequently reassigned to Judge Martinez-Olguin following Alsup's retirement. The settlement garnered remarkable participation, with authors and publishers filing claims for more than 91% of the approximately 480,000 eligible works. This participation rate significantly surpasses the 9% median typically observed in US consumer class actions, a statistic highlighted in a 2019 Federal Trade Commission report.
Lead plaintiff attorney Justin Nelson of Susman Godfrey underscored the historic nature of this claims rate, citing it as evidence of overwhelming class support. Under the proposed terms, each eligible title would receive at least US$3,000 before costs and fees, split between authors and publishers. Self-published authors and those who have regained their rights would be entitled to the full amount, deviating from the standard 50/50 split between publishers and authors for non-educational works.
Why it matters for Australian investors
The Anthropic case, while playing out in US courts, holds significant implications for Australian investors, particularly those with exposure to AI-focused technology companies or intellectual property (IP) intensive sectors. The ongoing legal battle underscores the complex and evolving landscape of copyright in the age of generative AI. Australian investors should recognise that similar legal challenges could arise within Australia's jurisdiction as AI adoption accelerates.
For investors holding shares in global AI firms, the financial and reputational risks associated with copyright infringement lawsuits are substantial. A US$1.5 billion settlement, or even more significant damages if a trial proceeds, can impact a company's valuation, profitability, and future growth prospects. Australian investors often access these global tech companies through exchange-traded funds (ETFs) or direct share purchases via local platforms that facilitate international trading, like those offered by major Australian brokers.
Furthermore, this case highlights the increasing scrutiny on how AI models are trained and the provenance of their data. Australian regulators, such as ASIC (Australian Securities and Investments Commission) and AUSTRAC (Australian Transaction Reports and Analysis Centre), while primarily focused on financial markets and anti-money laundering respectively, are part of a broader global push towards responsible and ethical AI development. While their direct involvement in copyright disputes is limited, the principles of accountability and legal compliance apply broadly across all industries, including the burgeoning AI sector in Australia.
Australian content creators, from authors to artists, also have a vested interest. The outcome of cases like Anthropic's could set precedents that influence intellectual property rights and compensation models globally, potentially impacting Australian creators' ability to protect their work and monetise its use by AI. This dynamic could, in turn, affect the investment landscape for local content-focused companies or platforms.
Impact on the AUD market
While the Anthropic settlement is a US-centric legal matter, its broader implications for the global digital economy could subtly affect the Australian dollar (AUD) market. A significant legal precedent set in the US regarding AI and copyright could influence international trade flows, specifically in digital services and intellectual property. Australia, as a net importer of technology and digital content from regions like the US, watches these developments closely.
Should there be a material shift in how AI companies operate or are compensated due to heightened licensing costs from such lawsuits, it could indirectly affect the operational costs for Australian businesses utilising these technologies. This might lead to higher prices for digital services, potentially impacting consumer spending or business investment, which could have a marginal effect on broader economic indicators watched by the Reserve Bank of Australia (RBA).
Moreover, investor sentiment towards the broader technology sector, particularly AI, can have flow-on effects. If major legal disputes like Anthropic's create uncertainty or reduce the perceived stability of tech investments, it could trigger shifts in global capital allocation. While the AUD's primary drivers remain commodity prices and interest rate differentials, a significant downturn in global tech investment sentiment could lead to investors seeking safe-haven assets, potentially affecting capital flows into or out of Australia.
However, it is crucial to note that direct and immediate impacts on the AUD market from this specific case are likely to be limited. The Australian economy is diverse, and the direct exposure of Australian financial institutions or the federal budget to the Anthropic lawsuit itself is negligible. Any effects would be indirect, emanating from changes in global tech sector confidence or regulatory frameworks that might eventually extend to Australia.
What to watch next
The immediate focus remains on Judge Martinez-Olguin's decision regarding the attorney fees and lead plaintiff payments. Her requests for more detail suggest a rigorous review process, and her final ruling will determine if the US$1.5 billion settlement can proceed. The judge's eventual sign-off (or lack thereof) will be a critical juncture, directly impacting the 120,000 authors awaiting compensation.
Another key development to monitor is the ongoing dissent among some authors. A group of more than 25 writers, including prominent novelists such as Dave Eggers and Vendela Vida, opted out of the settlement and filed a new lawsuit against Anthropic in California on May 13. This parallel legal action indicates that even if the current settlement secures final approval, Anthropic – and by extension, the AI industry – may face continued legal challenges from dissatisfied creators.
Separate lawsuits from other authors and publishers with similar claims are also proceeding, collectively highlighting the fragmented but persistent legal pressure on AI developers regarding copyright. These cases, alongside Anthropic's, will continue to shape the legal precedents for AI training data and intellectual property in the digital age. Australian investors and businesses should pay attention to these evolving legal frameworks, as they will undoubtedly influence the cost and methodologies of AI development and deployment globally.
Finally, the broader regulatory landscape will be crucial. As countries like Australia develop their own frameworks for AI governance, they will consider international legal developments. The outcomes of such high-profile cases could inform future Australian policy on AI ethics, data provenance, and copyright protection, potentially affecting local tech startups, digital content industries, and the broader economy.
Coins covered
Common questions
What is 'fair use' in the context of Australian copyright law, and how does it relate to AI training?
In Australia, the concept equivalent to 'fair use' in the US is 'fair dealing', which applies to specific purposes like research or study, criticism or review, news reporting, parody or satire, and judicial proceedings or professional advice. Unlike the broader US fair use doctrine, Australian fair dealing provisions are more prescriptive. The application of fair dealing to AI training in Australia is a developing area. Currently, using copyrighted material for AI training without permission may constitute infringement, unless it falls under an existing fair dealing exception or a specific licence. The Anthropic case highlights the global complexities, and Australian policymakers are watching these developments closely to inform future local legislation.
If I am an Australian author, how could the Anthropic settlement indirectly affect my rights or potential future compensation related to AI?
While the Anthropic settlement is a US case, its outcome could set an important precedent for how copyright holders globally are compensated when their works are used to train AI. If a large, successful settlement is finalised, it could encourage similar class-action lawsuits in other jurisdictions, including Australia, or influence industry-wide licensing agreements. Australian authors might see increased demand for licensing their works for AI training or better compensation models emerge, as AI companies seek to mitigate legal risks by legitimately acquiring data. This could affect the value of your intellectual property and future earning potential from digital uses.
Are there any Australian crypto exchanges or platforms that offer exposure to AI-related investments affected by legal cases like Anthropic's?
Australian crypto exchanges like CoinSpot, Independent Reserve, Swyftx, and BTC Markets primarily focus on trading cryptocurrencies directly. While some may offer tokens related to decentralised AI projects, they generally do not provide direct exposure to publicly traded AI company stocks or ETFs that would be directly impacted by the Anthropic lawsuit. Investors looking for exposure to traditional AI companies (like Anthropic, if it were publicly traded) would typically use traditional Australian stockbrokers or investment platforms that offer access to global equities and technology sector ETFs, rather than crypto exchanges.
Anthropic's record US$1.5bn AI copyright settlement hits a snag. CoinPulse AU analyses what this means for Australian investors and the future of AI IP rights
