Make your data readable by AI — and trustworthy
By adding simple, clean markup directly to your HTML or publishing standalone AODM documents, you make your content instantly understandable to AI systems, search engines, agents, summarizers, and generative tools.
AODM 1.2 goes further than describing what your data says. It records where each statement came from, how confident you are in it, when it is valid, and whether it was measured or inferred — the things a machine needs in order to use your data responsibly rather than just ingest it.
No heavy scripts. No performance hit. Just smarter data that works for you.
What AODM gives you
A small, stable core
Four knowledge primitives — entity, relationship,
fact, rule — plus three annotations. Small enough
that any system can implement it, stable enough to build on.
Provenance as standard
Every fact can answer “where did this come from?” and “how sure are we?” Sources and confidence are part of the format, not an afterthought bolted on by each consumer.
Knowledge that expires
Facts carry validity windows, so a figure that was true in 2024 and superseded in 2026 says so itself. Stale data stops quietly poisoning downstream results.
Measurements, not strings
A quantity is written as a number with a unit and a tolerance — or as a range — so consumers never have to parse “about 1.2°C” out of prose with a regular expression.
Known-false vs. unknown
AODM can state that something is not true, distinctly from having no information about it. A knowledge base that cannot tell those apart will answer questions it should decline.
Formally specified
An XML Schema, a JSON Schema, documented validation rules, and a conformance test suite. Anyone can build a compatible parser from the specification alone.
What it looks like
A measured fact, with its source, its confidence, the date it became valid, and a unit-tagged value:
<aodm:fact id="temp-rise" about="engine" valid-from="2026-01-15">
Temperature rise under sustained load.
<aodm:value number="1.2" unit="Cel" tolerance="0.1"/>
<aodm:source uri="https://example.com/report-24" retrieved="2026-01-15"/>
<aodm:confidence value="0.9"/>
</aodm:fact>The same document can be written as JSON, with an exact field-for-field mapping. Both forms are covered by a published schema.
Get started — it’s free and simple
- Download the package from the download page.
- Declare the namespace and add tags to your content — see the documentation for setup and worked examples.
- Check your markup with the online validator before you publish.
- Questions? Email support@nesaraq.com.
AODM: smarter markup for your site. A better foundation for AI.