Blog / AI and Data Teams

AI and Data Teams

AI Tools for Judges: What Belongs on the Bench and What Does Not

AI Tools for Judges: What Belongs on the Bench and What Does Not

Judges are already using AI. A Northwestern University study of 502 sampled federal judges found more than 60 percent of the 112 who responded had ...

From Search to Analysis: Working a Case File Without Reading Every Page

From Search to Analysis: Working a Case File Without Reading Every Page

AI case analysis for judges is easiest to understand as a spectrum with a hard stop at one end. At the near end is search, which nobody objects to. ...

Measuring Court Performance: Backlog, Adjournments, and Time to Disposition

Measuring Court Performance: Backlog, Adjournments, and Time to Disposition

Court performance backlog measurement is usually treated as a reporting obligation attached to the end of a program. The judiciaries that have ...

AI Translation of Judgments, Documents, and the Spoken Record

AI Translation of Judgments, Documents, and the Spoken Record

A judgment is delivered to the parties. If a party cannot read it, the delivery was formal rather than actual, and in jurisdictions where a ...

Can AI Transcription Be Trusted With the Official Court Record?

Can AI Transcription Be Trusted With the Official Court Record?

Asking whether AI transcription is accurate enough for court is the wrong question, because accuracy is not a single number and the record is not a ...

Searching a Court's Own Prior Rulings and Transcripts

Searching a Court's Own Prior Rulings and Transcripts

Ask a judge how comparable matters were handled before and you will get one of two answers. Either a recollection, which is honest and incomplete, or ...

Writing an AI Policy for a Court: Approval, Disclosure, Review, and Audit

Writing an AI Policy for a Court: Approval, Disclosure, Review, and Audit

Most courts writing a court AI policy are writing it late. That is not a criticism, it is the position nearly everyone is in, and it changes what the ...

Running LLMs On-Premises: The Realities Nobody Advertises

Running LLMs On-Premises: The Realities Nobody Advertises

Running an LLM on hardware you own takes about ten minutes. Download an open-weight model, start a runtime, send a prompt, watch the tokens arrive. ...

Model Portability: Keeping the Ability to Change Your Mind About AI

Model Portability: Keeping the Ability to Change Your Mind About AI

Ask whether you could move off your current AI provider and the answer is more encouraging than the question expects. Swapping which model answers a ...

Choosing an On-Prem Model: A Practical Evaluation Framework

Choosing an On-Prem Model: A Practical Evaluation Framework

Choosing which open-weight model to run on your own hardware is a measurement problem rather than a research problem. The decision that holds up is ...

Running Agentic AI Workflows on Infrastructure You Control

Running Agentic AI Workflows on Infrastructure You Control

Agentic AI runs on premise, and the interesting question is not whether the model can be served locally. That part is solved, and it sits inside the ...