Ding ding! this was never really a great vehicle for the broader questions of Copyright and AI Training — the Opinion did a good job of cutting through the smokescreen … THOMSON REUTERS vs ROSS INTELLIGENCE - 3rd Circuit opinion is now available ... case should also not be given Cert by #SCOTUS - again a bad vehicle for considering the broader questions ... chatgptiseatingtheworld.com/…

It is such an exciting weekend here at @Cambridge_Uni Judge Business School as it is the Annual Legal AI Hackathon (LLM x Law Hackathon 2026). Last night, I had the pleasure of presenting one of the Opening Keynotes for the Hackathon. Great to see many wonderful folks here today in very warm England !

🚨 BOOK ALERT -- Sign Up Now for *Free* Access ! 🚀 📚"Building AI-Native Professional Services Firms: Strategy, Economics, and Execution" 📖 By Daniel Katz, Michael Bommarito & Jillian Bommarito 🔥 🔥We’re releasing this book as a serialized book — full chapters delivered as weekly episodes — building up to the complete hardcover + digital release in early April 🥳 💻 Here’s what you get when you SIGN UP -- ✅ Every week you’ll receive the next full chapter straight to your inbox (FREE) ✅ Early access to the serialized release before anyone else ✅ You will be notified when pre-order is available 🏛️ Whether you run a law firm, consulting practice, accounting firm, or any professional services organization, this is your chance to start building your AI-native future one chapter at a time. ⚖️ Let’s build that future together. 🪏👷 👍 **Sign up** for the free weekly release here: theainativefirm.com/ ➡️ ⮕ Access the Prologue here: theainativefirm.com/chapters…

It was a real honor to deliver the Opening Keynote at the Royal Society’s AI & Law Conference last week in London ! Thanks to all of the organizers and fellow speakers for this important dialogue regarding the future of our field. Link to Event Here: royalsociety.org/science.../…

2025 Fantasy Draft is Complete ... ready for a great season ...

A market based read on the overall race here (in the medium term) ...

Announcing the release of the first in our Multipaper Series on the Kelvin Legal Large Language Model (KL3M). Our initial paper focuses on *TOKENIZATION* — the often-overlooked yet vital element of the LLM production process. Leading general-purpose tokenizers, such as those in Llama 3 and GPT-4o, often misrepresent critical domain-specific terms like 'EBITDA' and 'Fed. R. Civ. P.' by parsing them into individual characters or ignoring abbreviations. This inefficient and inaccurate tokenization can have far-reaching consequences, potentially impacting all downstream processes and increasing the likelihood of issues like hallucination. Our kl3m-004-128k-cased tokenizer offers delivering tangible benefits for those working professional documents in law and finance. We achieve a 9-17% reduction in token usage overall compared to GPT-4o and Llama3, despite having a smaller underlying vocabulary. This efficiency is particularly evident with specialized terminology, where our tokenizer uses up to 83% fewer tokens for legal terms and 39% fewer for financial terms. PAPER: papers.ssrn.com/sol3/papers.… GITHUB: github.com/alea-institute/kl…

anyway -- what's the deal with Delaware ...