Descrybe Open Connector gives legal organizations the freedom to combine their own expertise and processes with Descrybe’s legal intelligence.
BOSTON, July 27, 2026 — Descrybe, the legal technology company building trusted legal research and intelligence for wherever legal work happens, today announced the Descrybe Open Connector, a new initiative that enables…
Descrybe Legal Engine is now available through ChatGPT’s App Directory, bringing structured U.S. primary law, citation intelligence, treatment analysis, and source-grounded AI legal research into one of the world’s most widely used AI assistants.
Boston, MA — July 2, 2026 — Descrybe today announced Descrybe Legal Engine for ChatGPT, giving users…
Growth follows Descrybe’s collaboration with Anthropic on the Claude for the Legal Industry launch, where the company served as one of 22 launch collaborators.
Boston, Mass — June 29, 2026 — Descrybe, a purpose-built legal AI company, today announced that it has added more than 700 new paid subscribers in the…
The Descrybe Legal Engine gives Claude a verified legal data layer, making AI-assisted legal research and reasoning more powerful, reliable, and accessible.
Boston, MA — May 12, 2026 — Descrybe today launched the Descrybe Legal Engine for Claude, a connector that brings 300 million structured U.S. primary law records and the…
Boston, MA — March 25, 2026 — Descrybe, the purpose-built legal AI platform, has been selected as an approved e-resource by NELLCO Law Library Consortium, Inc. following a thorough committee review. The designation makes Descrybe available to NELLCO’s network of nearly 150 law libraries across 33 U.S. states, with international members…
DescrybeLM answered all 200 multistate bar exam questions correctly. ChatGPT, Claude, and Gemini each missed between 13 and 23 questions — and scored lower on legal reasoning quality across the board. Methodology and scoring rubric published.
BOSTON, MA — March 15, 2026
When AI gets a legal question wrong, the…
The AI legal research startup Descrybe today launched a “legal reasoning” product, DescrybeLM, that it says outperforms leading general-purpose AI models on a standardized legal reasoning benchmark — and it is publishing the methodology and scoring data to invite scrutiny.
The company also launched an all-new website that features the new product while…
With the legal research startup Casetext now kaput, having been killed off after its acquisition by Thomson Reuters, a newer legal research startup, Descrybe, has secured its spot in the widely used curriculum of the National Society for Legal Technology, joining established players including Lexis+, Westlaw and Bloomberg Law as one of…
Even free legal research platforms need to make money at some point, and so as Descrybe.ai today launches a paid upgrade, it is making it worth your while, offering a “Legal Research Toolkit” with a suite of features, including its own AI-driven citator, which it calls the Cytator.
As I wrote about Descrybe…
descrybe.ai, a free legal research platform that uses artificial intelligence to provide enhanced access to the law, today rolled out a major upgrade that includes a redesign of its platform, more nuanced and accurate search results, summaries of judicial opinions in Spanish as well as English, and simplified, plain-language summaries in both English and…
Every year, the International Academy of Digital Arts & Sciences (IADAS) hands out its Webby Awards, which honor excellence on the internet and recognize influential creators and brands.
There are actually two awards — The Webby Award selected by IADAS and The Webby People’s Voice Award, voted on by the online community of…
descrybe.ai, a year-old legal research startup focused on using artificial intelligence to provide free and easy access to court opinions, has completed its goal of creating AI-generated summaries of all available state supreme and appellate court opinions from throughout the United States.
descrybe.ai describes its mission as democratizing access to legal information and leveling…
Robert Ambrogi Blog
