Overview
- LawGeex’s AI beat human lawyers in reading contracts with speed and precision
- The AI posted an accuracy of 94%
- It managed to read a contract in 26 seconds, an average human lawyer took 92 minutes
Introduction:
In another stunning show, an AI technology outperformed human lawyers by reading through a contract with impressive accuracy. All the while doing this at an almost 200 times quicker speed.
This study was conducted by legal AI platform, LawGeex. It gave us another glimpse of AI’s the immense potential. The technology goes through contracts and flags any arguments or language that seems out of context.
Source: LawGeex
LawGeex made 20 highly experienced lawyers battle against a trained AI, for a task of spotting errors in a contract. Here are the results:
- Human lawyers took 92 minutes on average with 85% accuracy rate
- AI read the contract in 26 seconds with 94% accuracy
Not only this, the AI achieved a 100% accuracy rate in one particular contract while the best accuracy by a human lawyer was 96%. Incredible!
LawGeex’s deep learning AI was trained using tens of thousands of Non Disclosure Agreements (NDAs) to reach that level of speed and precision.
Our take on this
Lawyers don’t need to worry for their jobs (yet). Like any other technology, this has it’s limitations as well. Lawgeex thinks it should be used by the lawyers, rather than instead of lawyers.
This is not the first time AI has been used to help with legal expertise. As we reported earlier, Instraspexion’s AI can help companies predict if they’re about to be sued. Another case was when technology was used to predict the results of cases judged by the European Court of Human Rights.
But it just shows how far the field of Natural Language Processing has come, and how much potential it still has.
Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!
An avid reader and blogger who loves exploring the endless world of data science and artificial intelligence. Fascinated by the limitless applications of ML and AI; eager to learn and discover the depths of data science.
