Highlights –

  • Microsoft releases the Human-AI eXperience (HAX) toolkit.
  • The toolkit will support companies when creating AI systems that interact with people.
  • The HAX toolkit includes four components to assist teams throughout the user design process, from planning to testing.

Microsoft has released the Human-AI eXperience (HAX) Toolkit to support companies when creating human-centered artificial intelligence (AI) systems.

The toolkit is a set of practical tools that can help teams strategically create and responsibly implement best practices while developing artificial intelligence technologies that interact with people.

The HAX Toolkit comes as AI-infused products and services, such as virtual assistants, autocomplete, route planners, recommendations, and reminders, becoming popular and useful for many people. Also, these applications can do those things that aren’t helpful, like misunderstanding a voice command or misinterpret an image. There are some cases where AI systems can demonstrate disruptive behaviors or even cause harm.

This is one of the reasons why AI developers have moved for trustworthy AI guidance. Supporting responsible practices has typically focused on enhancing algorithms and models. Still, there is a critical need to make responsible AI resources available to those who build the applications that people use.

The HAX Toolkit helps design a practical tool that would translate human-AI interaction knowledge into actionable guidance.

Four components

The toolkit includes four components designed to assist teams throughout the user design process from planning to testing:

  • The Guidelines for Human-AI Interaction come with best practices for how AI applications should interact with people.
  • The HAX Workbook enables teams to prioritize guidelines and schedule the time and resources required to address high-priority items.
  • The HAX Design Patterns provides a flexible solution to handle common problems that come across while designing human-AI systems. The HAX Design Library consists of a searchable database of design patterns and implementation examples.
  • Teams make use of the HAX Playbook for identifying and planning unforeseen errors, such as a transcription error or false positive.

Expert’s view

Saleema Amershi, Senior Principal Research Manager at Microsoft Research, said: “Human-centeredness is really all about ensuring that what we build and how we build it begins and ends with people in mind. We started the HAX Toolkit to help AI creators take this approach when building AI technologies.”