About
Why we are building this in Dhaka
Because the legal AI that reached Bangladesh first was built for somewhere else, and the gap was not a translation problem.
01
The observation
Every practitioner we spoke to had already tried a general assistant, and every one of them had the same story: fluent, plausible, and quietly about a different country. The failure was never a missing feature. It was that the tool did not know it was out of its depth, and nothing on screen let you tell.
02
One design principle
An answer must point at the provision it relied on. That single constraint does more for safety than any amount of hedging language: it turns an unverifiable assertion into something a qualified person can check in under a minute, and it makes the model's ignorance visible instead of letting fluency paper over it.
03
What we are not
We are not a law firm and we are not trying to replace one. Bangladesh has roughly one advocate for every several thousand people, and the bottleneck in access to justice is time, not willingness. A tool that gives an advocate back their afternoon serves more people than one that tries to do without the advocate.
04
Where this goes
Chambers first, because that is where the hours are and where the standard is highest. If it holds up under people who argue for a living, it will hold up for everyone else.
Put it on a real matter
One month, one live file, your own drafts loaded. If your juniors do not get their hours back, you will know inside a fortnight.
AttorneyAI is a legal technology platform, not a law firm. It does not provide legal advice and no lawyer–client relationship is created.