Automate metadata with AI Builder in Microsoft 365
Artificial Intelligence (AI) is gaining traction at more companies and in our daily lives.
Here at Gravity Union, we’re excited about the potential for AI to help organizations and people - especially for tasks that are cumbersome, error-prone and not user friendly. While AI isn’t perfect, it can organize and tag neutral info like invoices more accurately and consistently than humans can. AI helps automate tasks that humans are poor at doing (or hate doing).
Microsoft is betting big on AI and building it into the Microsoft 365 platform. For example, AI models are now accessible in Power Platform with the AI Builder, and soon in SharePoint with Project Cortex.
This is huge! It puts AI into the hands of knowledge workers – it’s not only developers who get to play around with AI.
We recently explored using AI Builder to automate the extraction of metadata from a set of documents, such as invoices. In this video I show you how to train the model and get started with AI Builder:
What is AI Builder?
Microsoft built a new component into Power Platform called “AI Builder.” According to Microsoft, you use it:
“…to improve business performance by automating processes and predicting outcomes.
AI Builder is a turnkey solution that brings the power of AI through a point-and-click experience.”
How to train an AI Model
To use AI, knowledge workers start by training a model to improve a process.
What kind of process? Almost anything! In a work setting, we expect models to be initially trained for common documents management flows such as contracts, estimates, invoices, policies or a set of forms. The machine, with a little bit of training, will understand how to extract key information into metadata for use in a list, database, workflow or other business process.
What does all this mean for the future of work?
Knowledge workers are going to be ‘AI trainers’ rather than ‘data entry clerks’ in the future. You might spend a few minutes training AI to read a type of document so that the machine can do the repetitive chore of tagging and categorizing.
This is about having machines and people work together. It’s not (necessarily) about using machines to displace people. Machines do what they do best – which is sorting and shifting through volumes of data. Humans do what they do best – which is using their judgement to make a recommendation or make a selection from a set of options.
We hope that organizations take stock of processes that can be improved and focus on those improvements through AI, thus enabling people to make better decisions and deliver higher value.
Helpful Resources:
Dive deeper into this topic with these links:
What is the AI Builder and info about pricing
Use a form processing model inside a Power Automate flow (Microsoft help docs)
Collaborative Intelligence: Humans and AI Are Joining Forces (HBR article)
AI can be applied to the work you do today. Let us know if you want to learn more about this topic or if you have questions.
