Project summary:

A fluent legal answer can be misleading. It may pick the wrong jurisdiction, rely on outdated law, invent supporting authority or overlook facts needed to reach a conclusion. These mistakes is difficult for a non-specialist reader to spot. We, therefore, propose measuring them directly via qualified legal reviewers assessing answers under documented testing conditions.

What are this project's goals? How will you achieve them?

The study will measure whether supplying the correct law reduces serious errors and whether plausible law that does not apply misleads models. It will also measure how a common search tool changes performance. Results will help legal educators, public-interest organizations, courts and developers understand the specific limits of the tested systems.

The award would deliver the following:

• A protected 570-question evaluation set with verified legal sources and separate answer keys.

• Public methods, aggregate findings, uncertainty estimates, code and selected approved development examples.

• A comparable U.S. employment and civil-procedure analysis alongside the Caribbean study, where we can align model versions and testing conditions.

How will this funding be used?

Who is on your team? What's your track record on similar projects?

See websites: www.fahalways.com for commercial work done, www.fahalwaysfc.com for the details of the study

What are the most likely causes and outcomes if this project fails?

Lack of funding

How much money have you raised in the last 12 months, and from where?

No funding raised. Just starting