Law touches many aspects of our lives and businesses, yet accessing it is not always as easy as it should be.
A legal text may already be published. The information may exist somewhere. But finding the right text, determining whether it is still in force, tracing its amendments, understanding how it relates to other legislation, and then locating a relevant judicial precedent or reference can turn into a process that takes far more time and effort than it should.
That is where the idea for Bunud began.
We did not start with the question: How do we build a chatbot that answers legal questions?
We started with a simpler and deeper question:
Why is accessing and researching legal knowledge still so difficult?
The problem is not a lack of information
A significant amount of legal knowledge already exists.
There are laws, decrees, decisions, regulations, instructions, judicial precedents, and legal references. The problem is that the existence of this information does not necessarily make it easy to use.
The text may be buried inside a long PDF.
It may appear in an outdated search result.
You may find the original version of a law without its subsequent amendments.
Or you may need to move between several different sources just to build a complete picture of a single legal issue.
For a lawyer, this is more than a poor user experience. It is time spent finding information instead of analyzing it.
For a student, researcher, business owner, or ordinary individual, the barrier can be even greater: Where do you begin? What exactly should you search for? And can you actually rely on what you found?
The problem, then, is not that the law is unavailable.
The problem is that legal knowledge is still, in many cases, fragmented, difficult to search, and even harder to connect and understand.
Traditional search solves only part of the problem
Search engines transformed the way we access information, but they were not designed specifically to understand law.
You can search for the name of a law or the number of a decree and often find what you need quickly. But things become more difficult when you do not already know what you are looking for.
You may have a legal question, not the name of a statute.
You may be searching for a concept expressed in language different from the words you used.
You may need a group of interconnected legal texts rather than a single document.
This is where keyword-based search begins to show its limits.
Real legal research is not simply about finding a word inside a document. It is about understanding context, relationships, legislative history, and meaning.
Then came artificial intelligence
Large language models changed our expectations completely.
Suddenly, we could ask a question in natural language and receive, within seconds, a detailed answer that appears to have been written specifically for us.
That is a major advancement.
But in law, fluent language is not enough.
A convincing answer can still be wrong.
A cited legal provision may not exist at all.
The information may once have been correct but no longer be in force.
A model may confuse different legal systems or infer a rule that is not actually supported by the source.
In many fields, these errors may simply be inconvenient.
In law, they can be fundamental.
That is why we did not want to build an interface that places a language model behind a chat box and then asks the user to trust whatever it says.
In law, being able to produce an answer is not enough. You must also be able to know where that answer came from.
This is where Bunud came from
We want to make legal knowledge searchable, explorable, and understandable in a way that is closer to how people actually think.
You should be able to search for a legal text when you already know what it is.
But you should also be able to describe the legal issue when you do not know which text you need.
You should be able to move from a law to the amendments connected to it.
From one provision to related legislation.
From a question to sources you can open, read, and verify yourself.
And artificial intelligence should assist you in research and analysis rather than ask you to treat it as an independent source of truth.
That is why we do not think of Bunud as merely an online library of laws.
Nor do we think of it as merely an AI legal assistant.
What we are trying to build is a modern layer for accessing and working with legal knowledge.
Law is not a collection of isolated documents
One of the core principles behind Bunud is that law is not simply an archive of independent files.
Legislation changes.
One law amends another.
A provision may replace an earlier provision.
Legislation may be repealed in whole or in part.
A decision may derive its authority from a law, implementing instructions may be issued under it, and judicial precedent may shape how it is interpreted and applied.
If we display each document in isolation, we may be showing the texts, but we are not showing the full picture.
That is why, at Bunud, we care about the relationships between legal texts just as much as the texts themselves.
The goal is to gradually move legal research away from searching through documents and toward searching through an interconnected system of legal knowledge.
We want AI to be a tool for lawyers, not a replacement for them
One question has been repeated constantly since the rise of artificial intelligence: Will AI replace lawyers?
We think there is a more important question:
What should lawyers no longer have to waste their time doing?
A lawyer’s value should not lie in the ability to open dozens of files to locate a provision they already know exists somewhere.
Nor in copying parts of legislation into another document.
Nor in repeating research tasks that a machine can complete in seconds.
The real value lies in understanding the issue, constructing the argument, assessing risk, connecting facts to law, choosing a strategy, and taking professional responsibility for the decision.
If technology can reduce the time spent before reaching that stage, it does not diminish the lawyer’s value.
It allows the lawyer to focus more of their time on the work that actually requires them.
But Bunud is not only for lawyers
Lawyers are among the most important users we are building for, but they are not the only ones.
There is the student trying to understand a legal subject for the first time.
The researcher tracing related legislation.
The business trying to understand the legal framework surrounding a commercial issue.
And the individual who simply wants to know what the law says before deciding what to do next.
That does not mean everyone will be able to replace professional legal advice.
It means that accessing the underlying legal text and understanding its basic context should not have to be a complicated task.
Legal knowledge can become clearer without becoming less precise.
What are we trying to change?
We want it to feel natural to write a legal question the way you actually think about it, rather than the way a search engine expects you to phrase it.
To find relevant legal texts even when you do not already know their numbers.
To understand why a particular result appeared.
To be able to return to the original source.
And to see the law in context rather than as an isolated page.
We also want tools that have historically been available mainly to large institutions or specialized researchers to become accessible, in a simpler form, to a much broader group of people.
And we want all of this to work in Arabic from the beginning, rather than as a translation layer added later to a product designed for a different market.
This is only the beginning
Bunud will not be built in a day.
Legal knowledge is vast, constantly evolving, and complex by nature. And the more technology becomes involved in research and analysis, the more important accuracy, sources, and transparency become.
That is why we prefer to build this system gradually.
We add data.
We connect legal texts.
We improve search.
We observe how lawyers and researchers actually use the tools.
We learn where artificial intelligence can save meaningful time, and where it should step back and show the source instead of trying to provide a final answer.
Our vision, ultimately, is simple:
Make law easier to access, legal research more intelligent, and verification clearer.
That is why we built Bunud.
And this is only the beginning.
Blog
Back to Blog