AI × Law

Anthropic is facing a lawsuit alleging its AI systems infringe patents covering machine-learning technology inspired by neuroscience. The reported development matters because it puts a major AI company in the crosshairs of IP litigation tied directly to how its models are built.

A federal judge approved Anthropic’s $1.5 billion settlement in Bartz v. Anthropic, resolving claims over more than 482,000 books allegedly sourced from piracy sites. The article says the deal is the largest copyright recovery in U.S. history and underscores a key risk for AI companies: fair use may not protect unlawful acquisition or retention of training data.

The UK Civil Justice Council has issued an interim report and consultation paper asking whether formal court rules are needed for AI use by legal representatives and experts when preparing court documents. It proposes stricter disclosure for trial witness statements and expert reports, while saying no new rules are needed yet for statements of case or disclosure technology.

Harvey says Microsoft’s Corporate, External, and Legal Affairs group will use its legal AI platform, while Harvey will expand its use of Microsoft 365 and Microsoft 365 Copilot. The move matters because it shows a major in-house legal team at Microsoft adopting a third-party legal AI tool after a thorough evaluation, even as Microsoft builds its own legal agent.

Brazilian legal tech startup Jusfy closed a $15 million Series A led by Quona Capital, with backing from Thomson Reuters Ventures and The LegalTech Fund. Jusfy says its platform uses AI agents to support legal research, case management, intake and other workflow tools, and it plans to expand beyond Brazil and build more financial services for lawyers.

The Hamilton Lincoln Law Institute has joined a First Amendment lawsuit challenging a 2025 state law that limits artificial intelligence. This is a concrete AI-in-law development because it adds another legal player to litigation over how far states can regulate AI speech.

A Bentonville photographer was arrested over alleged deepfake child sexual abuse material, and a family whose 10-year-old daughter’s likeness was allegedly used is now suing xAI in federal court. The suit says Grok helped generate explicit images and seeks damages, deletion of the images, and safeguards against non-consensual sexualized content.

Law.com published the first part of a two-part series on how generative AI in patent drafting and prosecution could create privilege and work-product disputes. It also flags discovery strategies litigators may use when challenging patents that may have been drafted with AI assistance.

Willkie said it is partnering with OpenAI to develop AI solutions across its legal and business operations, including a firmwide rollout of ChatGPT Enterprise. The firm also expects to integrate OpenAI Codex into its AI development environment through Willkie Works, signaling a deeper push to build proprietary legal and business tools on top of frontier models.

A securities fraud class action has been filed against ZoomInfo and certain executives after the company cut 2026 revenue guidance and its stock fell nearly 33%. The suit alleges ZoomInfo misled investors about how its AI-integrated products were affecting customer retention, making it a notable AI-related disclosure case for public companies.

This update says Illinois biometric privacy claims continue to generate large class actions involving face scans, fingerprints, and voice data, with prior settlements including Facebook’s $650 million deal, Google’s $100 million settlement, TikTok’s $92 million settlement, and BNSF’s $75 million fund. It also notes the Illinois Supreme Court and a 2026 Seventh Circuit ruling that shaped damages and retroactivity, making this a significant watch item for companies using biometric or facial-recognition tools.

The article surveys the key U.S. copyright rules and lawsuits now shaping AI, including the Copyright Office’s position that AI-generated works lack human authorship and major cases against OpenAI, Microsoft, Anthropic, Stability AI, and Midjourney. For litigators, the signal is in the concrete trend: courts and the Copyright Office are testing where fair use ends, with a $1.5 billion Anthropic settlement and multiple class actions still moving forward.

Franklin County Law Library’s Ohio AI resource page compiles Ohio Supreme Court comments on judicial conduct rules covering AI. It says judges should not delegate decision-making to AI, should not use AI for independent fact investigation, and may need to disclose consulting AI in a pending matter—useful guidance for anyone litigating in Ohio courts.

OpenAI and Willkie Farr & Gallagher are collaborating to develop and deploy AI tools, with Willkie rolling out ChatGPT Enterprise across the firm and using OpenAI models to extend its Wendell Intelligence platform. The deal signals how Big Law is moving beyond off-the-shelf tools toward custom AI built on firm data.

In Schulte v. LinkedIn, Magistrate Judge Laurel Beeler denied plaintiffs’ objections to LinkedIn’s use of Relativity aiR for Review, including its use of search strings before AI review and requests for performance metrics. The ruling matters because the court treated AI-assisted document review as a routine TAR workflow and focused on proportionality, disclosure, and QC rather than on whether AI could make final responsiveness calls.

The article says generative AI is now embedded in daily legal operations and is creating new exposure across IP, privacy, employment, product liability, and bias claims. It also highlights sanctions over AI-generated fake citations, including a federal court imposing $110,000 in fines after a filing with 23 fabricated citations.

Vikk AI says it raised $4.2 million in financing, split between a $700,000 pre-seed and a $3.5 million seed round. The company says the money will speed product development for its consumer legal assistant, document tools, attorney discovery, and a new sponsored lawyer advertising product aimed at AI-powered legal journeys.

Louisiana will start enforcing new AI laws on Aug. 1 that expand criminal penalties for AI-generated sexual images, require disclosures for political ads that use AI-altered media, and set procedures for introducing AI-generated evidence in court. For litigators, the court-evidence rules are the most notable: parties must notify opponents before trial, and judges can hold admissibility hearings.

A Washington State Bar News piece says federal courts have sanctioned lawyers and even judges for AI-hallucinated citations, from a $110,000 penalty in Oregon to disqualification in Alabama. It highlights the WSBA’s Advisory Opinion 2025-05 and says the bar is considering proposed AI-related comments to the RPCs, underscoring growing discipline risk for lawyers who use AI without verification.

Arnold & Porter has created a new chief artificial intelligence officer role and promoted Roger Maeda into it. He will oversee developing, evaluating and implementing AI-powered tools for client work, signaling another Big Law firm putting formal leadership behind AI adoption.