Artificial Intelligence Markup Language

The open standard scripting language for chatbots

        
        
      

Background

We started with a mission. To create a simple language for crafting conversational agents that anyone could learn, interpret, or extend.

Our ongoing work continues to drive adoption of AIML as the open standard for chatbots, in the same way HTML is standard for the web.

We believe chatbots should be like websites (not mobile apps), and work seamlessly across all messaging and voice ecosystems.

Open

Not a black box. No obfuscating "AI" buzzwords. No solution provider lock-in. The opposite of ML/NLP APIs.

Multilingual

AIML can be written in almost any natural language. (Except Dolphin - we're still not sure how whale sounds work).

Flexible & Extensible

Integrate your APIs. Databases. CRM. Intent and Entity NLP APIs. Fork or write your own tags. The world is your oyster.

Easy to Learn

It takes a few days to learn the basics and a couple weeks to master, even if you're more of a copywriter than a coder.

XML-Based

Love tags? Hate tags? Either way, tags are a tried and true standard. (P.S. We are working on a simpler syntax....)

Market Tested

Seems stupid simple; works in production. Just ask 1M+ developers or look under the hood of the best chatbots.

Open-source Interpreters and Libraries

Visit us on Github for more resources. Third party software should be carefully vetted prior to use.

Looking for a hosted SaaS solution? Try Pandorabots: a free service for building & deploying AIML chatbots.

Pandorabots

    Member Role

  • Chair
  • Vice Chair
  • Spec Editor
  • Member
  • Member
  • Member
  • Member

Check out the latest version of the AIML specification, reference, and usage examples.

Documentation

Changelog

AIML 2.0 Revision 1.0.2.xx

March 23, 2018
  • Rich media Extensions: new tags support multimedia features common to many modern messaging systems and sanctioned as a standard by GSMA in RCS.

AIML 2.0 Revision 1.0.2.22

March 9, 2014
  • Zero+ Wildcards
  • Highest Priority Matching
  • Migration from XML attributes to tags
  • AIML Sets
  • AIML Maps
  • Loops
  • Local variables
  • Sraix
  • Denormalization
  • Pandorabots Extensions (date, request, response, unbound predicates, learn, learnf, explode)
  • OOB (Out of Band) Tags