AI × Law

Legal AI startup Legora is reportedly in talks to raise new capital at a valuation of $10 billion, according to Sifted. If confirmed, the mega-valuation would highlight surging investor demand for AI tools built specifically for professional services.

Facing victim lawsuits over Grok-generated child sexual abuse material, xAI is taking the rare approach of suing its own users for terms-of-service violations. The company is invoking an indemnification clause to force users who circumvented safety filters to reimburse its legal costs, setting up a major judicial test for AI platform liability.

The USPTO issued a disciplinary reprimand to a plaintiff's attorney after an AI-generated claim construction chart included erroneous citations to a patent's specification and figures. The case highlights increasing regulatory oversight and potential sanctions for attorneys who fail to verify AI-generated work product in IP litigation.

Mississippi officials have asked a federal appeals court to remove a district judge after he issued a preliminary injunction that allegedly relied on error-plagued AI-generated text. The state argues the judge's reliance on unverified generative AI output compromised the accuracy and fairness of the ruling.

San Francisco-based AI video-search SaaS Clipto raised $15 million in an all-equity round, bringing its post-money valuation to $250 million. The company reports $15 million in ARR, net-income profitability, and an expanding enterprise base that includes lawyers leveraging its local-first architecture for searchable media discovery.

Anthropic announced Enterprise Frontier Safeguards (EFS), allowing enterprise clients—including law firms and regulated industries—to run frontier AI models while storing logs entirely within their own cloud infrastructure. The architecture eliminates Anthropic's human review and third-party data retention, addressing critical legal privilege and client confidentiality requirements.

Moving beyond small pilots, AT&T's legal department created Legal Edge, a technology-driven center of expertise operating as an AI-first in-house practice. The team has expanded from 24 to roughly 35 people over the past six months to scale AI adoption across the organization.

India's Copyright Office ruled that Stephen Thaler, not his DABUS AI system, is the author of an AI-generated artwork under Indian law. The office found the work met the statutory originality requirement but rejected Thaler's effort to register the AI itself as the creator.

The California coordination proceeding (JCCP 5431) for ChatGPT product liability cases has reached 25 active or pending matters in San Francisco Superior Court, with Judge Ethan P. Schulman appointing plaintiffs' leadership. The consolidated litigation and a parallel suit by the Florida AG against OpenAI highlight how courts are beginning to process novel duty-of-care claims against AI developers.

The U.S. Court of Appeals for the Fifth Circuit questioned whether a case challenging a Mississippi anti-DEI law should be reassigned after a district court judge used AI to draft an error-riddled ruling. Judges raised concerns that the AI tool cited non-existent names and statements, questioning whether the hallucinated output infected the lower court's ultimate reasoning.

Private equity firm Silver Lake has become the largest shareholder in e-discovery provider Relativity following a deal that values the company at $3.6 billion. Relativity plans to use the investment to continue expanding its legal technology and e-discovery offerings.

Draftwise has released Legal Ontology by Draftwise, a platform designed to map law firms' institutional knowledge into a connected intelligence layer. The tool helps firms analyze client history, deals, relationships, and legal obligations across disparate systems.

Fireworks AI announced the general availability of its Training API and Fireworks Lab. As part of the release, the company revealed that legal AI startup Harvey post-trained Kimi K3 using asynchronous reinforcement learning on the platform, achieving a 19.7% all-pass score on the LAB benchmark compared to 10.8% for the base model.

UK-founded legal tech startups raised a record £189 million across 2025, marking a 35% increase from the previous year according to LawtechUK data. Generative AI tools targeting document and contract management led investment activity, while recent major exits like Microsoft's acquisition of Robin AI signal growing enterprise appetite for legal tech infrastructure.

A prominent panel of judges and legal experts published "The Judicious Judge’s Guide to Generative AI and LLMs," providing ethical and technical guardrails for judicial chambers. The guide outlines permissible uses like summarizing briefs and checking citations, while warning against delegating substantive judicial reasoning or uploading confidential court filings to public models.

Federal Court Justice Michael Lee has issued novel orders requiring McDonald's and the union plaintiff to consult tech experts on using AI ahead of mediation and trial in a major underpayment class action. The judge aims to use the multi-million dollar dispute as a testing ground to significantly reduce massive case expenses.

Clio is embedding its Vincent AI platform directly into Clio Operate, allowing AI to access the structured history, documents, and workflows of active legal matters. The integration combines legal research and drafting with real-time matter context, with general availability scheduled for November 2026.

Blue Voice, an AI platform trained on local laws, department guidelines, and municipal ordinances, has emerged from stealth with $6 million in funding led by SignalFire and Las Olas VC. Founded by a Harvard Law dropout, the platform acts as an instant reference tool for 225 police agencies to quickly verify legal standards in the field.

California-based Encore Law Group has outlined its official AI policy, permitting matter-specific generative AI use strictly under attorney oversight while prohibiting consumer-tier tools. Notably, the policy warns clients that inputting case details or attorney advice into AI chatbots risks waiving attorney-client privilege.

AI legal tool developers Harvey, Clio, and Legora have roughly doubled their combined annual recurring revenue over the past year, reaching $1 billion.