LegisGate API · planned
The same determination, planned as an API.
LegisGate works out which laws apply to an AI use case and keeps the result as one structured record, each obligation cited to its provision. A licensed LegisGate API that delivers that record to your own product is planned and would be provided by agreement. Today, two public data feeds are free to read.
What the record holds
One record for each AI use case, with the citation behind every line.
The platform stores the result of each determination as a structured record. The planned API would deliver that record, not a summary of it and not prose about the law.
The laws that apply, and the ones that do not
Each law that reaches the AI use case, and each law that was checked and does not apply, with the condition that was not met. A question that went unanswered is recorded as pending, never silently as a no.
Obligations with their citations
Each obligation names the provision it comes from, with the source link and the date it takes effect.
Classifications
The EU AI Act scope and tier for the AI use case, and the factors behind it.
Where it came from
Each record is stamped with the version of the Meridian it was determined against.
{
"evaluated_count": 313,
"corpus_revision": 16608,
"applies": [
{
"instrument_name": "GDPR",
"jurisdiction": "European Union",
"citation": "Regulation (EU) 2016/679 (General Data Protection Regulation) art. 22",
"title": "Give a meaningful human review to the AI tool's automated decisions"
},
{
"instrument_name": "NYC Local Law 144",
"jurisdiction": "New York City",
"citation": "N.Y.C. Admin. Code § 20-871(a)(1)",
"title": "§ 20-871(a)(1) Bias audit conducted no more than one year before use"
}
],
"ruled_out": [
{
"instrument_name": "Colorado AI Act",
"reason_category": "out_of_footprint",
"reason": "This law does not cover any of the jurisdictions you declared."
}
],
"pending": [
{
"instrument_name": "UK IDTA",
"field_key": "transfer_mechanism_type",
"reason": "The question was not answered, or the answer was not known."
}
]
}This is an excerpt of the summary the platform shows for the demonstration AI use case, the Applicant Screening Assistant, with each list shortened. It illustrates the shape of the record. It is not the response of a licensed API, because that API does not exist yet.
What this is not. LegisGate produces regulatory intelligence and assessment-preparation materials. Nothing we produce is legal advice, a legal opinion, a certification, or a determination of compliance, and no document we deliver discharges a legal obligation on its own. Your counsel determines sufficiency. A determination tells your customer what attaches. It does not tell them they have met the law, and your product should not imply that it does.
What goes in
The answers to the dynamic intake form.
A determination starts from coded answers: the AI system type, the data subjects, whether there are automated decisions, the jurisdictions and about ninety more for the demonstration AI use case. The dynamic intake form is a questionnaire that asks only what your earlier answers make relevant.
The inputs are answers, not free text
Fixed rules, applied by software, compare the answers with who each law covers in the LegisGate Meridian™. No language model takes part in the determination.
What was ruled out is part of the answer
The laws that were checked and did not attach are recorded with the condition that was not met. The absence of an obligation has value only if you can show that it was looked for.
A missing answer is never silently a no
Where a law was kept in because a question went unanswered, the record says so.
Available today
Two public data feeds, free to read.
These are live now. They carry dates and counts from the Meridian, not a determination for your AI use case.
The Statutory Deadline Register
GET /api/deadline-register returns the next 24 months of statutory dates as JSON. GET /api/deadline-register/rss returns the same dates as a feed.
Meridian coverage
GET /api/meridian/coverage returns the Meridian in numbers: laws, jurisdictions and obligations.
Inside the platform
The same engine produces the determination you see in your workspace. Your first five AI use cases are free, with everything included, so you can see a determination before you decide anything about the API.
If you want to build on it
Tell us what you would build.
The licensed API is planned and is not built yet. Its endpoints, credentials, schema and service levels are not published, because they do not exist yet. If you would use the determination inside your product, tell us how; it shapes what is built first.
What you would use it for
The product, the AI use cases it touches, the jurisdictions and the volume.
The terms
A license would be provided by agreement, including what your interface must not imply about the result.
The integration
How your product sends answers and receives the record would be agreed and built for the license.
Build on the determination
Tell us what you would build on it.
If you provide software that touches AI use cases, questions about deployer obligations reach your support team whether or not you have an answer. A licensed LegisGate API is planned. Talk to us, and start free to see the determination first.