Judge Pushes Back, Lawyers Cut Fee in $1.5B Anthropic Settlement
Lawyers involved in a landmark $1.5 billion copyright settlement with AI company Anthropic have significantly reduced their request for legal fees following criticism from both the court and the defendant, highlighting tensions over attorney compensation in major class-action cases.
The case stems from a lawsuit filed by authors and publishers who accused Anthropic of using pirated books to train its artificial intelligence models without permission. The dispute became one of the most closely watched legal battles in the AI industry, as it raised fundamental questions about copyright, data usage, and the boundaries of “fair use” in training large language models.
To resolve the claims, Anthropic agreed to a $1.5 billion settlement—widely considered the largest copyright class-action settlement in history. Under the agreement, the company also committed to destroying infringing datasets and compensating authors, with payouts estimated at more than $3,000 per copyrighted work.
However, controversy quickly arose over how much the lawyers representing the plaintiffs should be paid. Initially, the lead law firms—Susman Godfrey and Lieff Cabraser—requested approximately $300 million in legal fees, which represented about 20% of the total settlement fund. A portion of this amount, around $75 million, was intended to be shared with three additional law firms that were not formally appointed as class counsel.
This proposal drew sharp criticism from both Anthropic and the judiciary. Judges questioned whether it was appropriate to allocate such a large share of the settlement to attorneys, particularly when some of the firms seeking compensation were not officially designated to represent the entire class of plaintiffs. U.S. District Judge William Alsup, who previously oversaw the case, expressed concerns about the structure of the fee arrangement, emphasizing that class counsel cannot simply delegate their responsibilities to other firms while still seeking payment for their work.
In response to this pushback, the plaintiffs’ lawyers revised their request downward to $187.5 million, or 12.5% of the settlement. Importantly, the new proposal excludes fees for the three additional law firms, focusing solely on compensating the officially recognized class counsel. The attorneys stated that this revised figure more accurately reflects the work performed on behalf of the class.
Despite the reduction, questions remain about how the excluded firms will be compensated, if at all. Some of those firms have argued that they played a meaningful role in negotiating the settlement and preparing the case for trial. Still, the court’s stance suggests a stricter interpretation of how legal fees should be distributed in class-action lawsuits, prioritizing transparency and accountability.
The case is now under the supervision of U.S. District Judge Araceli Martinez-Olguin, who will decide whether to grant final approval to both the settlement and the revised fee request. A hearing is scheduled to determine whether the agreement is fair to all parties involved, particularly the authors and publishers who stand to benefit from the compensation fund.
Beyond the immediate dispute over legal fees, the case has broader implications for the rapidly evolving field of artificial intelligence. It represents one of the first major legal resolutions involving claims that AI systems were trained on copyrighted material obtained through unauthorized means. The outcome could influence how future cases are handled and how companies approach the sourcing of training data.
The controversy also highlights ongoing debates about attorney compensation in large settlements. Critics often argue that excessive legal fees can reduce the amount of money available to plaintiffs, while supporters contend that complex litigation requires significant resources and expertise that justify substantial compensation.
Ultimately, the decision to reduce the fee request may help smooth the path toward final approval of the settlement. However, it also underscores the scrutiny that courts are applying to both tech companies and the lawyers who challenge them, particularly in high-stakes cases involving emerging technologies.
⚖️ Key Legal Outcomes
- Lawyers reduced their fee request from $300 million to $187.5 million.
- The revised request excludes compensation for three non-class counsel law firms.
- The $1.5 billion settlement with Anthropic remains pending final court approval.
- Judicial criticism played a key role in forcing the reduction.
- A federal judge will determine whether both the settlement and fees are fair.
📌 Why It Matters
- Sets a benchmark for legal fees in massive AI-related class actions.
- Highlights growing judicial scrutiny over attorney compensation.
- Impacts how much compensation ultimately reaches affected authors.
- Signals stricter enforcement of rules governing class counsel roles.
- Shapes future legal battles over AI training and copyright use.
