This week, AI’s collision with intellectual property law dominated headlines, as Sony and Warner Music’s sweeping lawsuit against Anthropic signals an escalating legal war over training data that could reshape the entire AI industry. Meanwhile, courts in Australia and the US are actively pioneering AI as a procedural tool inside litigation itself — a sign that judges, not just firms, are now driving AI adoption in legal proceedings.
Top story: Sony Music and Warner Chappell filed a landmark federal lawsuit against Anthropic alleging mass copyright theft of thousands of songs to train Claude, seeking billions in statutory damages.
Sony and Warner Sue Anthropic Over Mass AI Training Theft
TechCrunch · Legal
Sony Music Publishing and Warner Chappell Music filed a federal lawsuit against Anthropic and its co-founders, alleging the company conducted a ‘brazen campaign of illegally torrenting, scraping, and downloading copyrighted works’ to train its Claude AI models. The publishers claim approximately 20,000 songs were pirated — including works by The Beatles and Taylor Swift — and are seeking up to $150,000 per infringed work, a figure that could run into billions. The case is particularly significant for legal teams because it follows Anthropic’s landmark $1.5 billion settlement with authors, signalling that courts are increasingly unwilling to treat AI training data acquisition as legally neutral.
Australian Judge Orders Parties to Use AI in $100M McDonald’s Case
Proctor (Queensland Law Society) · Legal
In a landmark procedural move, Australian Federal Court Justice Michael Lee formally ordered parties in a $100 million McDonald’s underpayment class action to investigate and report back to the court on whether AI could be used to collate evidence, prepare chronologies, and manage documents. Parties have until 6 November 2026 to produce a protocol under the court’s existing Generative AI Practice Note. This is one of the first instances globally of a judge mandating AI exploration as a case management obligation, setting a precedent that could normalise AI-assisted litigation workflows across common law jurisdictions including the UK.
https://www.qlsproctor.com.au/2026/09/judge-orders-parties-to-probe-use-of-ai-to-prepare-evidence/
Manifest AI Law Firm Raises $60M at $750M Valuation
Lupl Blog · Finance
AI-native law firm Manifest has raised $60 million at a $750 million valuation, according to a mid-2026 review of the AI-native law firm landscape. The raise is part of a broader wave of capital flowing into firms that aim to replace, not just augment, traditional legal workflows with AI agents — a model that puts direct pressure on conventional law firm economics. The funding milestone signals that investors are now betting on AI-native practices as structural competitors to Big Law, not merely as legal tech vendors.