Whilst computer algorithms can process information, they can’t easily understand what that information means unless the information is structured. Words and sentences in a document are hard for a computer to understand because they do not always fit a structure which implies a specific meaning. Equally, meaning can be expressed using many combinations of words and sentences which makes the learning process harder for computers. Legislate and Legal Schema are working to solve this problem.
Why are contracts difficult for machines to process?
Contracts have historically been hard to process by machines because they contain lots of words and agreements of the same type can be written in many shapes and forms. Even humans find legalese difficult to process and clauses are in most cases tailored to the specific requirements of the parties which means the underlying concepts the algorithms are trying to identify are hard to generalise. As a result, contracts have mostly required humans in the loop to process the information they contain and act on it.
What is Legal Schema?
Legal Schema (UKLS) is an initiative backed by LawTech UK, described in this white paper, and officially launched in June 2021 to help machines process contract information more easily by giving structure to contract data. A schema provides a framework for describing the structure of data so that machines can understand the underlying concepts. For example, a contract’s parties, their respective obligations and dates can be expressed in a structured way so that they are machine readable. A simple example of structured data to describe a company is provided below:
"name": "Legislate Technologies",
"streetAddress": "2 Littlegate street",
"postalCode": "OX1 1QT",
What is Structured data?
Structured data is used in the same way by Google to understand content on a page. For example it can be used to mark up Frequently asked questions, reviews and other concepts which in turn can be linked to other known entities or displayed in a more relevant way in search results. For example, our how legislate works page is marked up as a “How To” which means that it will display in search results as a set of instructions:
Since a digital contract has a similar structure as a web-page, it can be marked up with structured data. Legal Schema is proposing a general framework for structuring the data of contracts in the UK. Making the information of a contract machine readable means that smart contracts can transfer their structured information to other systems such as a CRM or payments platform without a human in the loop. This could in turn trigger events automatically or when certain conditions are met. For example, a smart contract could execute a payment when a delivery is confirmed.
Legal Schema’s proposed structured data is primarily aimed at operationalising the performance of the assets described in a contract. Legal Schema give an example of a dematerialised bill of lading which describes the asset which is being transferred and the conditions of the transaction. The transactions of Legal Schema contracts could ultimately be performed using blockchain-based smart contracts. Connecting dematerialised contracts to external systems can reduce fraud and friction in commercial transactions.
What contract data can be structured?
Providing structure to the assets of a contract is very valuable but only covers a fraction of the data present in a contract. Legal terms can and should be structured as well to unlock the full potential of machine readable contracts. An agreement contains legal terms that are structured and organised in an interconnected way. For example, a non-disclosure agreement’s structure will be affected by whether one or both parties are disclosing confidential information as well as the nature of the confidential information. Structuring the terms of a contract and modelling the relationships in an ontology means that smart contract terms can be configured into a coherent contract without a human in a loop. Structuring the legal terms of a contract also makes it possible to query the legal terms of a contract, such as a party’s obligations and restrictions. For example, with structured contracts, an organisation could quickly identify the agreements in which they are disclosing confidential information, and filter them by the type of information disclosed.
How can legal data be structured?
Legislate is pioneering the use of knowledge graphs (United States Patent No. 11,087,219) to structure both the asset and legal data of contracts. Knowledge graphs are a type of database which stores data as relations as opposed to tables. Knowledge graphs are ideal for navigating highly interconnected data and finding patterns. Moreover, knowledge graphs can be taught concepts using ontologies which is why they are currently used by Google to process structured data. Legislate’s knowledge graph models the legal terms and their relationships which means that it knows how they are connected and how they can be combined together to form a valid contract. As a result, Legislate is helping the unlawyered create contracts and easily access their contract information post-signature.
What are the benefits of structuring legal data?
As technology becomes ubiquitous, it is important that information in any form is machine readable. Legal Schema is proposing a schema for structuring data in contracts which is helping dematerialise contract assets and in turn reduce friction in transactions. Legislate is proposing a knowledge graph approach (United States Patent No. 11,087,219) to structuring contract data which is richer and enables machine readable legal terms. This unlocks the full potential of structured contracts which can now be assembled and searched automatically. Finally, structured contracts reduce the friction in the creation process and makes safe contracting accessible to the unlawyered. Thanks to Legislate the unlawyered can create and manage smart contracts by themselves.
Legislate is an early-stage legal technology start-up which allows large landlords and small businesses to easily create, sign and manage contracts on their own terms. Legislate’s knowledge graph approach (United States Patent No. 11,087,219) is unlocking the full potential of contract data. Legislate’s team marries technical and legal expertise to create a painless, and unique, contracting experience for its users. Legislate is backed by Parkwalk Advisors, Perivoli Innovations and angel investors.