Skip to content
K = 0.731Subscribe

Legora Tests Legal Agents in Simulated Environments to Protect Client Data

2 reports 1 reporting sources Updated 4 hours ago

Understand the event

AI overview

Legora builds and tests legal agents for lawyers without reading their work traces, CTO Jacob Lauritzen said, because those records belong to law firm clients. The company uses simulated evaluation environments modeled on customer work, paraphrased traces, and a process that narrows failures by type. It runs its systems on Azure. Lauritzen estimated that models already perform about 60% of legal work well.

Generated by AI from reports · Updated 3 hours ago

Latest development11/3 07:01 (UTC+8)
Legora CTO Jacob Lauritzen on Building Legal AI AgentsLatent Space

Read next

More changes in today’s briefingOpenAI’s updated Model Spec explained through its five core values

Report timeline

Follow the reports to explore different sides of the event.

11/3
  1. Latent Space
    Legora CTO Jacob Lauritzen on Building Legal AI Agents

    In a Latent Space interview, Legora CTO Jacob Lauritzen describes how the legal AI company builds agents for lawyers after raising $550 million at an $8.8 billion valuation. He explains how three engineers and three lawyers build simulated eval environments because customer data is confidential, why Legora runs on Azure, and how legal engineers and forward-deployed engineers split client work.

  2. Latent Space
    Legora CTO Jacob Lauritzen on Building Legal Agents Without Seeing Client Data

    In a Latent Space interview, Legora CTO Jacob Lauritzen described how the legal tech company builds and tests agents when it cannot read traces of what they do for lawyers, because the data belongs to law firm clients. Legora uses simulated test environments modeled on customer work, paraphrased traces and failure-type narrowing, and runs everything on Azure. Lauritzen estimated models already do about 60% of legal work well.

Keep up with AI, one email a day

Model releases, price changes and compute developments, with original sources. Free, confirmed by email, and easy to unsubscribe. Read a past issue

Event heat trend

Current within comparable coverage
9
Peak within comparable coverage
911/3 09:00 (UTC+8)
24-hour change
–

Trends compare the same continuously observed participants, which may cover fewer sources than current heat. Point or click to inspect hourly heat, or use left and right arrow keys.